——————— BRARIES ””5“ \\\\\\\\\\\\\\\\\\\\\l\\\l\l\\\\l\\Rl\\\\\\\\\\\\\\\\\\\\\\\\ LIBRARY Michigan State . University This is to certify that the dissertation entitled PLANTINGA AND THE RATIONALITY OF THEISM presented by Thomas John Burke, Jr. has been accepted towards fulfillment of the requirements for _P_hLD_. degree in W Wig/fl Date _MQ;L_3_O_1_19_89_— rofessor MSU is an Affirmative Action/Eq ual Oppormm'Iy Institution 0-12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MSU Is An Affirmative Action/Equal Opportunity Institution c:\clre\ddedtnpm3—p.1 14. a1 mpa :irWITTNGA AND THE RATIONALITY OF THEISM by Thomas J. Burke, Jr. A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Philosophy 1989 RecentlY: “Vim thesis that belief that is, given the foundationalism, he belief in God could existed. This study claim does not abet t tries to show that alternative to both this alternative, i neither evidential i m Properly basic 1 It is argued Plantinga’s weak fou too easily defeasibl that a case can be liefs epistemicall foundationalism. i n God were proper' (000 I49.)( ABSTRACT PLANTINGA AND THE RATIONALITY OF THEISM BY Thomas J. Burke, Jr. Recently, Alvin Plantinga has put forth the provocative thesis that belief in God might well be "properly basic". That is, given the general epistemological position of foundationalism, he has argued that under certain conditions belief in God could be rational even if no evidence for God existed. This study attempts to demonstrate that Plantinga’s claim does not abet the epistemic fortunes of Theism, and then tries to show that what is needed is an epistemological alternative to both foundationalism and coherentism. Given this alternative, it is concluded, belief in God will be neither evidential in the "classical foundationalist" sense nor properly basic in any foundationalist sense. It is argued first that the sorts of beliefs which Plantinga’s weak foundationalism allows as properly basic are too easily defeasible to function as foundational beliefs, and that a case can be made for limiting proper basicality to beliefs epistemically stronger than those countenanced by weak foundationalism. Second, it is contended that even if belief in God were properly basic in some situations, as such it would too easily d of epistemic value The root Pr0b3 in agreement with is untenable and account for epistt Plantinga’s week he considers prot coherentist schen indicates that 51 evidential, but a I conclude 1 structure is nec According to thi highly theoretic: its justificati importance withi as viable as its would too easily degenerate into an evidential belief to be of epistemic value. The root problem, it is claimed, is that both foundation— alism and its primary rival, coherentism, are inadequate to account for all the types of beliefs which we have. I argue in agreement with Plantinga that "classical foundationalism" is untenable and that coherentism cannot satisfactorily account for epistemic warrant. But the similarities between Plantinga’s weak foundationalist schema for justifying what he considers properly basic beliefs and Laurence BonJour’s coherentist schema for justifying his "observation beliefs" indicates that such beliefs are neither properly basic nor evidential, but a third type. I conclude that an alternative construal of our noetic structure is necessary and attempt to outline such a view. According to this proposal, I claim, belief in God plays a highly theoretical role in our noetic structure. Therefore, its justification requires demonstrating its essential importance within a theoretical framework which is at least as viable as its nontheistic rivals. Copyright by THOMAS JOHN BURKE, JR. 1989 Appreciation following without v completed. First, sending me unpubli and for his kind Second, to Hillsda which I doubtless think, write and Elizabeth for he Prodding. Fourth their time to sex above all, I won] his direction dur 0111‘ Conversations always encouragir Criticisms, inaccuracies, am '1 “ every aSpect. ACKNOWLEDGEMENTS Appreciation and thanks must be expressed to the following without whose help this study would never have been completed. First, to Alvin Plantinga for his kindness in sending me unpublished works of his which proved invaluable, and for his kind permission to use quotations from them. Second, to Hillsdale College for the sabbatical leave without which I doubtless would never have had sufficient time to think, write and ultimately finish. Third, to my wife Elizabeth for her support, encouragement, and occasional prodding. Fourth, to the members of my committee who gave of their time to scrutinize the finished product. Finally and above all, I would like to thank Professor Richard Hall for his direction during the writing of this dissertation. Both our conversations together and his manuscript notations were always encouraging and instructive, and without his insights, criticisms, and deftness at ferreting out inconsistencies, inaccuracies, and ambiguities, this study would be much poorer in every aspect. ii I. INTRODUCTION - 1.1 The Comte 1,11 Set 1.12 ThE Phi 1,2 contempol Anxiety 1,3 Evaluatii II. THE ATTACK ON 11.1 Evident 11.2 Foundat 11.3 The Prc 11.31 11.32 11.33 11.34 11.4 Conclus III. PROPER PROPEl IV. THEISM, PROF II. III. IV. TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . I.l The Contemporary Epistemic Anxiety over Theism I.11 Setting the stage . . . . . . . . . . . I.12 Theism in the Context of Contemporary Philosophy . . . . . . . . . . . . . . . I.2 Contemporary Theistic Cures for Epistemic Anxiety . . . . . . . . . . . . . . . . . . . I.3 Evaluating the Basicality of Theism . . . . . THE ATTACK ON FOUNDATIONALISM . . . . . . . . . . . II.1 Evidentialism . . . . . . . . . . . . . . . II.2 Foundationalism . . . . . . . . . . . . . . II.3 The Proper Basicality of Theism . . . . . . . II. 31 The Great Pumpkin Objection . . . . . . II.32 The Ground of Belief in God . . . . . II.33 Is Argument Irrelevant to Basic Belief in God? . . . . . . . . . . . II.34 Fideism . . . . . . . . . . . . . . . . II.4 Conclusion . . . . . . . . . . . . . . . . . . PROPER PROPERLY BASIC BELIEFS . . . . . . . . . . . III.l Plantinga on Proper Basicality . . . . . III.2 Problems with Plantinga Properly Basic Beliefs . . . . . . . . . . . . . . . III.3 Proper Properly Basic Beliefs . . . . . . . . THEISM, PROPER BASICALITY, AND JUSTIFICATION . . . IV.1 Plantinga Properly Basic Beliefs and Rational Theism . . . . . . . . . . . . . . . . . . . . IV.11 Belief in God as an Innate Belief . . . IV.12 Belief in God as an Authority Belief IV.13 Belief in God as an Observation Belief IV.14 Theism and Direct Observation of God IV. 15 Conclusions . . . . . . . IV.2 Design and Warrant . . . . . . . . . . IV.3 Warranted Belief or Warranted Believing . . IV.4 Internalist Constraints and Type- -Token Difficulties . . . . . IV.5 The Justification of Theism’ s Basicality . . IV.6 Conclusion . . . . . . . . . . . . . . . . . iii 1 2 3 12 15 22 26 27 34 43 44 51 51 52 53 57 58 71 85 94 95 97 102 104 111 112 115 122 127 130 133 V. AUTHORITY BELIl v.1 Authority v.2 Authority v.3 Authority v.4 ConclusiOI VI. JUSTIFYING BAS V1.1 Planting Warrant V1.11 T s V1.12 8 v1.13 w V1.14 p v1.15 5 VI.2 BonJourj v1.21 c V1.22 I 1 VI'3 Conclusj v11. RATIONAL Nonr VII'l Nonba VII.1 VII.1 V. AUTHORITY BELIEFS . . . . . . . . . . . . . . . . 135 v.1 Authority Beliefs and Proper Basicality . . . 137 v.2 Authority Beliefs and CF Justification . . . 145 v.3 Authority Beliefs and Rationality . . . . . . 168 v.4 Conclusion . . . . . . . . . . . . . . . . . 169 VI. JUSTIFYING BASIC BELIEFS . . . . . . . . . . . . . 172 VI.1 Plantinga on Evidence, Basicality, and Warrant . . . . . . . . . . . . . . . . . . 176 VI.11 The Structure of a Rational Noetic Structure . . . . . . . . . . . . . 176 VI.12 Basis and Support . . . . . . . . . . 178 VI.13 Warrant . . . . . . . . . . . . . . 181 VI. 14 Plantinga’ s Rejection of Coherentism 192 VI. 15 Summary . . . . . . . . . 195 VI.2 BonJourian Coherentist Observation Beliefs . 195 VI.21 Observation Beliefs and Coherentism . 196 VI.22 BonJour’s Justification of Observation Beliefs . . . . . . . . . . . . . . . 203 VI.3 Conclusion . . . . . . . . . . . . . . . . . 212 VII. RATIONAL NOETIC STRUCTURES . . . . . . . . . . . 215 VII.1 Nonbasic Nonevidential Beliefs . . . . . . 217 VII.11 Forming Nonevidential Nonbasic Beliefs . . . . . . . . . . . 219 VII.12 Warrant Criteria for Nonevidential Nonbasic Beliefs . . . . . . 227 VII.13 Nonevidential Nonbasic Beliefs and Justification . . . . . . . . . . 237 VII.2 Properly Basic Beliefs . . . . . . . . . 242 VII. 21 The Epistemic Status of Basic Beliefs . . . . . . . . . 243 VII.22 Properly Basic Beliefs and Justification . . . . . . . . . 249 VII.3 Implications for Foundationalism . . . . . 252 VII.4 Implications for the Justification of Theism . . . . . . . . . . . . . . . . . . 254 VIII. CONCLUSION . . . . . . . . . . . . . . . . . . . 255 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 261 iv The twentieth a1 Theism which ha: history. While against the ruli response in the pa the middle of this belief were clearl the halls of phi academe. Belief i eXpression of emot content at all . has once again be We position. to the labors of ,- sone of the cent CHAPTER I INTRODUCTION The twentieth century has not been kind to the tradition- al Theism which has dominated the past 1500 years of Western history. While sceptical voices have often been raised against the ruling religious beliefs, the intellectual response in the past has been adequate to sustain faith. By the middle of this century, however, the forces of theistic belief were clearly in retreat, being all but vanquished from the halls of philosophical discourse in most of Western academe. Belief in God was considered irrational, simply the expression of emotional attachments, and.without any cognitive content at all. The philosophical fortunes of Theism have seen a drastic improvement in the past 30 years, however, and belief in God has once again become a permissible if not yet fully respec- table position. Much of this new found acceptability is due to the labors of Alvin Plantinga, whose efforts at dispelling some of the central and most powerful objections to Theism have met with a good deal of success. In the past 10 years, he has directed his labors towards erecting a positive case for the rationality of Theism. His approach has not been to demonstrate the truth of Theism, but its rationality, in essence, arguing th The central fo claim, for if Plat significant step ft of Theism. Broadly epistemic gains su successful, and if In the remain t0 Place Plantingav first briefly Outl which has placed Tl summarize the majo; ing or rejecting summary of the var this state of aft shall OUT-line the of argumentatioh t I. 1 THE CONTEMPO] 2 essence, arguing that belief in God can be properly basic.l The central focus of this study will be directed at this claim, for if Plantinga is right, this claim would be a significant step forward in the intellectual revitalization of Theism. Broadly speaking, I shall attempt to evaluate the epistemic gains such a claim would provide Theism were it successful, and if it is not successful, to show why. In the remainder of this introduction, I shall attempt to place Plantinga’s claim in its historical context. I will first briefly outline the general epistemological background which has placed Theism in its present difficulties, and then summarize the major contemporary reasons for either withhold- ing or rejecting theistic belief. Section II will give a summary of the various responses which Theists have given to this state of affairs, including Plantinga’s. Finally, I shall outline the general theses of this study and the line of argumentation to be developed in behalf of their defence. I.1 THE CONTEMPORARY EPISTEMIC ANXIETY OVER THEISM How is it that Theism has come upon such hard times? A telling of the complete story would easily encompass a few lMore accurately, he has not argued that belief actually is properly basic, (although he clearly believes that to be the case), but that it’s proper basicality is possibly possible. Even if it were properly basic, it would only be so under certain conditions, but Plantinga has not presented arguments to date that this actually is the case. Rather, he has presented arguments which, if successful, would show that belief in God could possibly be guoperly basic for some persons under appropriate conditions. volumes tracing t science, and theolo and would require 1 fields. The part c study is, of coursi philosophy from the portion of the tal than this author pc by noting the overa Placing Plantinga’ Theism in context . LII Settin the Philosophy of critical analYSlS SUCh differs only philosophy ' AS SI 3 volumes tracing the intellectual history of philosophy, science, and theology over the past five or six hundred years and would require the talents of not a few experts in those fields. The part of the story that is of importance to this study is, of course, the epistemological history of Western philosophy from the inception of the Modern period. Even this portion of the tale would require more space and expertise than this author possesses, but I must at least set the stage by noting the overall setting of Western epistemology, thereby placing Plantinga’s argument for the proper basicality of Theism in context. I.11 Setting the Stage Philosophy of Religion as currently practiced is just the critical analysis of religious beliefs and concepts, and as such differs only in subject. matter from other areas of philosophy. As such, it attempts to determine the meaning, consiStency, and validity of these beliefs and concepts. Unlike some other areas of philosophy, however, philosophy of religion also concerns itself with arguments for and against the existence of that entity which gives to religion its raison d’etre, viz., God. Philosophy of science does not doubt the existence of its central objects, scientific method and theories, nor philosophy of mind the exitence of minds, although of course what minds are might well be debated. The first task of philosophy of religion, however, is directed at the question of whether or not the object of religious practice actually e God does not exist, religious beliefs a philosophically rel reality (or at lea implications for hi that makes critica while, indeed, even cal reality of ti threaten the very e even granting ant would still have ex and epistemic stat enterprise. If, h phil°5°PhY 0f reli< sociology and p51 importance. 4 practice actually exists. Surely, if it could be shown that God does not exist, philosophers would concern themselves with religious beliefs and practices only insofar as they revealed philosophically relevant aspects of human nature. It is the reality (or at least the possible reality) of God and the implications for human existence which that reality entails that makes critical reflection on religious beliefs worth- while, indeed, even urgent. Even the debate over the ontolog- cal reality of theoretical entities in science does not threaten the very existence of philosophy of science because, even granting antirealism, scientific theories themselves would still have existence and the analysis of their structure and epistemic status would remain a worthwhile philosophical enterprise. If, however, atheism were to prove indubitable, philosophy of religion would appear at best quixotic, although sociology and psychology of religion would retain their importance. The existence of God (or the gods) and the importance of the proper service of God. were taken for' granted. by all societies for centuries, even millennia, before anyone thought to question in a systematic and critical way the validity of belief and worship. likely, it was the Greeks who first turned a critical eye towards this universal human preoccupa— tion.2 The context for our current reflexion, however, probably has its proper origin in Medieval Scholasticism, for 2In the West, at least. it was during that unswerving religiot carefully articulat intense philosophi< constructed the f0 various problems up of religion has for of God's existence divine foreknowledg of the soul, the n religious language, have their roots 1] develOPEd under t1 Course, Previous accomplishments, 1 along with a host addressed by the s the very method 0 clarity and Preci: Moreover, the has and proposed 3011 deve10ped by the S‘ were ignorant of s scholastics had i 5 it was during that heyday of heady rationalism coupled with unswerving religious conviction that a fully developed and carefully articulated concept of God was first subjected to intense philosophical scrutiny. Moreover, the scholastics constructed the foundational context out of which grew the various problems upon which.modern and contemporary philosophy of religion has focused. For example, not only the question of God’s existence, but also such topics as free will and divine foreknowledge, divine and human action, the immortality of the soul, the nature of the divine being, the meaning of religious language, and the theistic foundations of ethics all have their roots in the concepts of God and nature which were developed under the impetus of Medieval Scholasticism. Of course, previous thinkers contributed to the scholastic accomplishments, particularly St. Augustine and Aristotle, along with a host of lesser lights, and many of the problems addressed by the scholastics had been debated previously, but the very method of scholasticism gave to these questions a clarity and precision which they seldom had attained before. Moreover, the basic categories in which the formulations of and proposed solutions to these problems are framed were developed by the scholastics--although later generations often were ignorant of some of the very helpful conceptual tools the scholastics had invented. In general, the scholastics were firm believers that faith and reason are complementary and that the man or woman of faith has no reason to fear the questionings of reason, for reason itself is ca able answer to 8V9} is able to demonstr any threat posed b adverse evidence ar Theistic framework. For that schoo the most influenti rational defense of Thomas at the outs Aristotelian categ< which assertedly by about the world a deduce, using neces 019195, the exists revelation, fulfil Of the Church were guarantors of the 1 to this rational ; they could show t difficulties whic‘ without violating 6 reason itself is capable of showing that faith has a reason- able answer to every reasonable question. Moreover, reason is able to demonstrate with certainty that God exists, and so any threat posed by apparent conceptual inconsistencies or adverse evidence are merely problems to be solved within the Theistic framework. For that school of scholasticism which eventually became the most influential, the Thomist, the cornerstone of this rational defense of Theism was the "five ways" outlined by St. Thomas at the outset of his Summa Theologica. Using basic Aristotelian categories, St. Thomas sketched five arguments which assertedly began with some universally admitted "fact" about the world and showed that from that fact one could deduce, using necessarily true philosophical and logical prin- ciples, the existence of God. Of course, the evidences of revelation, fulfilled prophecy, miracles, and the authority of the Church were also considered support for the rationality of Theism, but the philosophical proofs, not being subject to doubt by anyone capable of following them, were the ultimate guarantors of the rationality of faith. Moreover, in addition to this rational foundation for faith, the scholastics felt they could show that any and all of the various conceptual difficulties which might be conjured up could be resolved without violating any rational principles. The sixteenth century brought not only theological and humanistic onslaughts against the fortress of scholasticism, but the onset of a philosophical approach to questions of metaphysics and e epistemic assumptio The assurance with v of the external w epistemic abilities less, the ways in held common sense a at least a psycho approaches to knowi this time seem not It is part 01 those eras in whic) the world happen a scepticism. The "m most egregious ex, Western world EXpei knowledge from the its philosophers c °f scepticism to scepticism of Node 7 metaphysics and epistemology which simply rejected the epistemic assumptions upon which scholasticism was founded. The assurance with which the scholastics accepted the reality of the external world and the trustworthiness of human epistemic abilities was simply considered untenable. Doubt- less, the ways in which empirical science was showing long held common sense and "scientific" beliefs to be mistaken had at least: a psychological effect. which. made the sceptical approaches to knowledge and reality that we find arising at this time seem not only plausible, but necessary. It is part of the irony of intellectual history that those eras in which great gains are made in our knowledge of the world happen also to be accompanied by various forms of scepticism. The "Modern" period in philosophy is perhaps the most egregious example of this phenomenon, for while the Western world experienced a monumental explosion in empirical knowledge from the sixteenth through the nineteenth centuries, its philosophers created some of the most intractable forms of scepticisni to date. Indeed, as time progresses, the scepticism of Modern philosophy seems to become more and more endemic to the philosophical enterprise itself, a fundamental assumption of philosophy per se, culminating in Kant’s "Copernican Revolution" according to which our empirical "knowledge" tells us nothing about the world itself, but only about the world of our experience; and from this experience we obviously cannot infer anything about the nature of "objec— tive", "external" reality. The ironY in surprising when we between sceptical 0 current ideas are 51 noetic condition is appearances is a p knowledge, for unle why it is that we motive, incentive, "common sense" or mother of technical intellectual inven‘ Moreover, dour by which better and Descartes was not of the world is, by we “many do haw “him We acquire i1 external world b. indubitable belief all, what better w 8 The irony in this intellectual history becomes less surprising when we reflect upon the symbiotic relationship between sceptical concerns and advances in knowledge. Unless current ideas are seriously questioned, no improvement in our noetic condition is possible. In a sense, scepticism towards appearances is a prerequisite for advancement in empirical knowledge, for unless we question what it is we perceive or why it is that we perceive it as we do, we have neither motive, incentive, nor rationale for going beyond either "common sense" or inherited theory. If necessity is the mother of technical invention, doubt is clearly the father of intellectual invention. Moreover, doubt is not usually an end itself, but a means by which better and more secure understanding can be obtained. Descartes was not seeking to show how tenuous our knowledge of the world is, but rather a way by which one could show that we actually do have knowledge of the world despite the way in which we acquire it. His attempt to validate knowledge of the external world by means of deducing that validity from indubitable beliefs is simply a logical first attempt. After all, what better way to prove that one's beliefs are true than showing that they follow from beliefs about which one cannot be wrong? Indeed, the major thrust of modern epistemology has been towards discovering "foundational" beliefs fronlwhich our nonfoundational beliefs can be inferred and showing what sorts of inferences properly provide the connections between them. In fact, this reliance upon foundational beliefs for philosophical Pr°°f modern philosohy i differences between evaluations of what philosophy, not in of security that st In addition, m connected with cum and related physic; human epistemic sy. reasons for sceptiw doubts. Both th counter-arguments . secure noetic four connecting that fc (lists have tried indubitable and t belief-worthiness some knmedge and are nevertheless y In a broad SE "foundationalist" 9 philosophical proofs is one of the important ways in which modern philosohy is continuous with scholasticism. The differences between them at this point lie in the differing evaluations of what counts as a warranted starting point for philosophy, not in their respective evaluations of the sort of security that starting point must provide. In addition, modern epistemological theories are closely connected with current scientific understandings of perception and related physical phenomena. The new findings about how human epistemic systems function has given rise to both new reasons for sceptical doubts and possible responses to those doubts. Both the sceptical arguments and the proposed counter-arguments assume that what the dilemma requires is a secure noetic foundation and a relatively secure means of connecting that foundation with other beliefs. Epistemolo— gists have tried to show that certain of our beliefs are indubitable and that given these indubitable beliefs, the belief-worthiness of other beliefs follows, so that we do have some knowledge and so that beliefs of lesser epistemic stature are nevertheless belief-worthy. In a broad sense, this is true not only of the strictly "foundationalist" epistemologies one finds in the seventeenth and eighteenth centuries, but of the "coherentist" systems produced in the nineteenth and twentieth centuries as well. While the former ground nonfoundational beliefs in foundation- al ones, the latter ground all beliefs in the "system as a whole". Although. coherentists deny' that any’ beliefs are individually indubi' secure beliefs by 1 beliefs, the resul within that system belief would be tot are less likely t< centrality in the s Locke might see thc experience as most beliefs which are 1 in the system as tr these appoaches hav and coherentism agr think we know we d< Given this gel 10 individually indubitable, they nevertheless argue that one can secure beliefs by their interconnected relations with other beliefs, the resultant system then lending to all beliefs within that system their epistemic security. Even though no belief would be totally secure in such a system, clearly some are less likely to be given up than others due to their centrality in the system. So while foundationalists such as Locke might see those beliefs which arise from our immediate experience as most certain, a coherentist would see those beliefs which are most centrally supported by other beliefs in the system as those which are the least revisable. While these appoaches have significant differences, foundationalism and coherentism agree in holding that much of what we commonly think we know we do know. The sceptic then criticizes this. Given this general epistemological context, the sceptic usually sets out to show either that the proposed justifica- tion for" belief, whether incorrigible beliefs or a ‘total system of beliefs, is not sufficient grounds for claiming knowledge and/or that from such a ground one cannot adequately justify other beliefs. The thrust of modern epistemology, then, has been to find adequate justifications for knowledge claims, but those justifications have been increasingly difficult to find . But if there is controvery over whether there are any foundational beliefs, and if so what sorts of beliefs are or could be foundational, there is virtual unanimity that one needs evidence for any nonfoundational belief if it is to be belief-worthy. De} must be traceable simply must be a co evidental support, must have evidentia must be justified. to foundational be also the case in a tions to other beli 1y secure the beli Thus, the 501 most Western philc as dependent upon is Wt a foundati i“corrigibly true dence. Theism cj Iigible, so its 1 evidence Which car evidenCe needs at scientific endeav. t0 the Same lex scientific propos method seem to he ll belief-worthy. Debate may rage over whether that evidence must be traceable to foundationl beliefs or whether there simply must be a coherent set of other beliefs which provides evidental support, but regardless, nonfoundational beliefs must have evidential support. Moreover, this support itself must be justified. Clearly, if one can trace a belief back to foundational beliefs, its support is justified, but it is also the case in a coherentist system that a belief’s rela- tions to other beliefs must be such that the latter justified— ly secure the belief-worthiness of the former. Thus, the sort of general epistemic outlook common to most Western philosophy sees rational acceptance of beliefs as dependent upon the epistemic status of that belief. If it is not a foundational belief, i.e., one self-evidently or incorrigibly true, it must be supported by sufficient evi— dence. Theism clearly is neither self-evident nor incor- rigible, so its rationality would seem to depend upon the evidence which can be mustered in its behalf. Moreover, this evidence needs at least to approach the kinds of demands which scientific endeavors require if belief in God is to lay claim to the same level of rational commitment and trust as scientific propositions do, for science and the scientific method seem to have uncovered at least some of the necessary conditions for the satisfactoriness of evidence. 1,12 Theism in th Given the framw ogy, it is not SL1] various attacks. I against the validit notions, it would b raised against re: scarcely an area of been subject to sex on the part of see There have be ticism and two for critical historica the ninetheenth a justification for : accuracy and the authenticity and . minor historical versions of a St( related facts disc 12 I.12 Theism in the Context of Contemporary Philosophy Given the framework.of Modern and Contemporary epistemol- ogy, it is not surprising that Theism has been subject to various attacks. If strong sceptical doubts have been raised against the validity of our common sense and older scientific notions, it would be surprising if those same doubts were not raised against religious beliefs also. Indeed, there is scarcely an area of human learning and knowledge which has not been subject to severe attack by virtue of epistemic scruples on the part of sceptics. There have been two primary reasons for religious agnos- ticism and two for religious disbelief. The application of critical historical methods to the Old and New Testaments in the ninetheenth and twentieth centuries has provided the justification for sceptical attitudes towards their historical accuracy and the traditional claims of their religious authenticity and theological truth. It was not simply that minor historical discrepancies were noted between different versions of a story in the Bible or between the Bible and related facts discovered by archeology. While the latter have by and large only been uncovered recently, apparent inconsis— tencies within the Bible had been noted since the Patristic era, and the history of biblical studies is rife with various attempts to harmonize them. Conflicts with nonbiblical sources, although new and important, were not fatal, for on the one hand it was possible the Hebrew record of the events was the more accurate account and on the other hand it was possible new infOU What was more destri of the religious t of the development theses about relic ticular which were theology. The syr worked out in such in post-Tridentine was completely und The legitimac was further reinfc Theism. Bertranc anticipated reply about Theism encap God, not enough e right, None of th cally withstoOd c the logical diff drawback that com off and haVe the 13 possible new information would reconcile the discrepancies.3 What was more destructive to confidence in the trustworthiness of the religious tradition was the wholesale reconstruction of the development of that tradition coupled with explanatory theses about religion in general and Christianity in par— ticular which were incompatible with the truth of traditional theology. The synthesis of faith and reason that had been worked out in such detail in the Middle Ages and then redone lepost-Tridentine Catholic and Protestant systematic theology was completely undone by the various theories promoted. The legitimacy, for some the duty, of affirming disbelief was further reinforced by the apparant lack of evidence for Theism. Bertrand Russell’s well known remark about his anticipated reply at the Judgment if he should prove wrong about Theism encapsulates this point: "Not enough evidence, God, not enough evidence". To many, this reply seems just right. None of the traditional theistic proofs has unequivo— cally withstood criticism, and probabilistic arguments face the logical difficulties in any such argument plus the drawback that competing materialistic theories seem no worse off and have the added advantage of being more parsimonious 3Indeed, this often has turned out to be the case, as Fundamentalist literature delights in pointing out. See, e.g., Howard F. Vos, Genesis and Archaeology (Chicago: Moody Press, 1963) and Gleason L. Archer, Jr. A Survey of Old Testament Introduction (Chicago: Moody Press, 1964). See also K. A. Kitchen, Ancient Orient and Old Testament (Chicago: Inter-Varsity Press, 1966) and the works of the international- 1y renowned archeologist W. F. Albright for arguments suppor— ting the general historical reliability of the Old Testament by nonfundamentalists. ontologicallY' The positive < factors: (a) the a‘ existence and nati particular which I inconsistent with f of the new theorit have only created which could be wor odds with theology ments for theolog offered not only a religion, but ale behavior previous] The most sev. philosophical cir. evil' Although 1 l4 ontologically. The positive case for atheism has rested largely on two factors: (a) the availability of theories explaining both the existence and nature of the world in general and man in particular" which either do not, require or' are implicitly inconsistent with Theism and (b) the existence of evil. Some of the new theories the scientific revolution has produced have only created difficulties for theology, difficulties which could be worked out in due course. But others are at odds with theology per se and were often proffered as replace- ments for theological beliefs. Freudianism, for example, offered not only an alternative explanation for the origin of religion, but also a new 'theory to explain the sorts of behavior previously identified as "sinful". The most severe and popular argument against Theism in philosophical circles, however, has been the argument from evil. Although the God of Theism is not logically incom— patible with evil, even the amount of evil we have in this world, on the surface at least, it seems highly unlikely that such a God has created this sort of universe and continues to maintain it in its present condition. Any decent God, espec- ially the sort of God Theists believe in, it is maintained, would never have allowed such a state of affairs to develop in the first place. To many, the amount and intensity of evil in this world seems good reason for holding not merely to agnosticism, but to full—fledged atheism. Even though many of these theories have no more factual basis than the thec and despite the spa upon which criticai tionai theology ha: defensive. It is I'll how show its viabi and theories which thesis of faith and plenty of reason ‘ Although we have m the universe" whicr is well enough cc belief, many would approach to date c< 1-2 Wow There have b. misfortunes of T] characteristic of 15 basis than the theological ideas they have sought to replace and despite the sparcity of archeological and historical data upon which critical theories of the Bible are based, tradi— tional theology has clearly been placed on the intellectual defensive. It is now up to the theological tradition to some- how show its viability given the mass of new facts, ideas, and theories which abound on all fronts. Lacking a new syn- thesis of faith and reason, religious scepticism seems to have plenty of reason to remain sceptical of religious claims. Although we have no nonreligious or materialistic "theory of the universe" which adequately accounts for all experience and is well enough confirmed to stifle any and all religious belief, many would argue that the success of the materialistic approach to date coupled with the difficulties theology faces make the former the odds on favorite. I.2 CONTEMPORARY THEISTIC CURES FOR EPISTEMIC ANXIETY There have been five major responses to the epistemic misfortunes of Theism in the past century. The first, characteristic of main-line Protestantism in the first part of the twentieth century, has been labeled "Modernism" or theological liberalismn For the most part, "Modernists" simply yielded as much to secularism as possible, eventually rejecting most all traditional dogmas and reinterpreting religious faith and religious beliefs in such a way that they often hardly differed in substance from nonreligious views. Indeed, extremes were sometimes reached which clearly transformed faith of God" theologies More stalwart qians and philoso such as Garrigou—l Marcel, courageou suPport to a revit 0f the Thomistic to show that the g Superior to its C( phical World rema to Roman Catholic Neo-Thomism fOr outside the Catho Phy was hardly a1 due to the reliar 16 transformed faith into nonfaith, such as the so—called "Death of God" theologies of the mid-sixties. More stalwart was the response of Roman Catholic theolo- gians and philosophers. Under the inspiration of thinkers such as Garrigou-Lagrange and later Etienne Gilson and Gabriel Marcel, courageous efforts were made to give philosophical support to a revitalized and revamped Thomism. Reformulations of the Thomistic proofs were offered and attempts were made to show that the general philosophical approach of Thomism is superior to its contemporary rivals. But most of the philoso- phical world remained unconvinced, and while those committed to Roman Catholicism frequently relied upon the resources of Neo—Thomism for the rational foundations of their faith, outside the Catholic Church the tide of anti—theistic philoso- phy was hardly affected at all. This, of course, is largely due to the reliance of Thomism upon Aristotelean concepts and principles. The modern and contemporary nonCatholic philoso- phical community simply rejects those assumptions. Another response to the increasingly secularist world View of which most Western philosophy was a part, was that of biblical literalism. Particularly characteristic of American Fundamentalism, herculean attempts were and still are being made to show that the Bible actually is errorless, that archeological data rightly understood supports all facets of the biblical picture of the ancient world, and that "true science" is compatible with the Genesis account of creation and man's early history. Philosophical and historical arguments, some 0 viability of seculz sophy, science and for the truth of t] characteristic of theorists, not by persuasion. Agai effect upon the ph A fourth resy spearheaded by Ka liberalism in whj provided nothing r the pulpit. He The°l°gy and sour simply rejecting seemed Clearly es known as "Neo-Or "sin", "fallennes traditional theol in? the literal t 01d and New Testa] l7 arguments, some of theni quite sophisticated,‘ against the viability of secularist views of human life, morality, philo- sophy, science and society in general supplement the arguments for the truth of the Bible. Moreover, while this response is characteristic of "6-day creationists" and "young earth" theorists, not by any means are all in this camp of that persuasion. Again, however, this position has had little effect upon the philosophical community, for obvious reasons. A fourth response to the challenge posed to Theism was spearheaded by Karl Barth. Barth rejected the theological liberalism in which he was educated because, he felt, it provided nothing of substance by which to nurture souls from the pulpit. He reached back into the roots of Reformed Theology and sought to present its central dogmas without simply rejecting those items of contemporary belief which seemed clearly established. In general, the position became known as "Neo-Orthodoxy", for while it preserved talk of "sin", "fallenness", "redemption", and the vast majority of traditional theological categories, it did so without suppos- ing the literal truth of either all the stories found in the Old and New Testaments or of what were considered the mythical construals of traditional theological concepts. In regard to epistemology, however, Nee-Orthodoxy simply rejected the demand for evidence for or proof of God’s existence. Barth in particular was especially adamant in his ‘See, for example, Carl F. H. Henry’s, God, Revelation and Authority, 6 Vol. (Waco, Texax: Word Books, 1983). rejection 0f unatu to "prove" GOd/s e1 attempt to do so V of human pride and under its own powe efforts Barth deci astance might he obviously not the empirically and dominated 1511910“1 against the phil actually had devz Barthians.6 Inde were vanquished i for evidence seem demands were theo In contrast i epistemic demands has pursued a st: strating the rati \ 5The title 0 18 rejection of "natural theology". Not only was it impossible to "prove" God’s existence by means of human reason, even the attempt to do so was viewed as simply an outrageous example of human pride and hubris, an attempt by mankind to bring God under its own power and to control its own destiny. To such efforts Barth declared a resounding "Nein".5 But while such a stance might have substantial theological merit, it was obviously not the sort of response which would impress the empirically and positivistically minded philosophers who dominated Anglo-American philosophy, and their arguments against the philosophical tenability of Barth’s position actually had devastating effects upon a number of younger Barthians.6 Indeed, in more cases than not, Barth’s views were vanquished in the philosophical arena where the demand for evidence seemed much more rational than the cry that such demands were theologically ill advised. In contrast to these largely theological responses to the epistemic demands being placed upon Theism, Alvin Plantinga has pursued a strictly philosophical approach toward demon— strating the rationality of Theism. Moreover, he employs the 5The title of the well known article in which he rejects Emil Brunner’s call for a mild and adulterated form of natural theology. 6For example, some of those who accepted Barth’s position early in their careers were instrumental in developing the "Death of God" theology under pressure from Logical Positivist attacks on the meaningfulness of religious language when, as with Barth, the latter is construed as completely beyond empirical verification. See Paul M. van Buren, The Secular Meaning of the Gospel (New York: The MacMillan Company, 1963). basic methods 0f gaining an immedi persuasion- USing modal 109iC and p01 and precision 0f dismissed out of h Piantinqa’s 0 God’s existence it The negative aspe demonstrating the philosophical argu while the positive in God is rational Logical Positivisn verifiability crii incoherent", its ( belief in God irr 19 basic methods of the analytic tradition itself, thereby gaining an immediate hearing among philosophers of that persuasion. Using such recent philosophical developments as modal logic and possible world semantics, the sophistication and precision of his arguments prevents them from being dismissed out of hand. Plantinga’s original forays were not attempts to prove God’s existence itself, but to show that Theism is rational. The negative aspect of this program has been directed at demonstrating the untenability of epistemological views and philosophical arguments which would make Theism irrational, while the positive aspect has sought to exhibit that belief in God is rational. Thus, he joined the broad attack against Logical Positivism, basically arguing that since the famed verifiability criterion of meaning was "self-referentially incoherent", its claim that "God exists" is meaningless and belief in God irrational ought simply be ignored.7 At the same time, he also presented arguments designed to prove the compatibility of Theism with evil.8 However, he has not merely been content to show that Theism had not been disproved; he has also attempted to show that theistic belief is rationally warranted. To this end, he has argued that the teleological argument is as good as any _ 7See, God and Other Minds, (Ithaca, New York: Cornell University Press, 1967), pp. 156—168. _ 8See, The Nature of Necessity, (Oxford: Oxford Univer— Sity Press, 1974) and God, Freedom, and Evil, (New York: Harper & Row, 1974). known argument for in other minds is claimed that the I as rational as it believe as to dis} Plantinga's Rather than try tc God exists, he onl in God is not irr: convert, but to Philosophy from rational belief, ticism, then one , cannot thereby be Cally based SOlut me as much pLin other perspectiVe 20 known argument for the existence of other minds, so if belief in other minds is rational, so is Theism.’ Likewise, he has claimed that the key premise in the ontological argument is as rational as its negation, so it is just as rational to believe as to disbelieve in God.10 Plantinga’s strategy, it should be noted, is novel. Rather than try to rationally coerce belief by "proving" that God exists, he only attempts to show that one who does believe in God is not irrational on that account. His aim is not to convert, but to make a case for the legitimacy of doing philosophy from a theistic perspective. If Theism is a rational belief, as rational in fact as atheism or agnos— ticism, then one who begins philosophy from that perspective cannot thereby be consigned to philosophic oblivion. Theisti- cally based solutions to common philosophical problems will have as much prima facie plausibility as those offered from other perspectives and will deserve the same attention. The spectre of Logical Positivism and its challenge to Theism’s rationality has passed from the scene, but other epistemologically based objections remain. The lack of any cogent proof of God's existence or any unambiguous evidence favoring belief in God has also been used to support an anti- theistic argument. Rational belief, it has been argued, should only be lent to beliefs for which there is evidence, 9God and Other Minds, pp. 187—271. ”God, Freedom and Evil, pp. 85-112. and since there rejected. That disbelieve in unic them, so one she evidence for His the rationality c disable this argu Plantinga ha; ing evidence for in God plausibly Conceivably, it n as we have seen, reQUire eVidence rational justific deriVes from 0the which give rise this possibility tlaliSmll grounde 21 and since there is no evidence for Theism, it should be rejected. That is, just as one has a rational duty' to disbelieve in unicorns because there is no evidence at all for them, so one should disbelieve in God since there is no evidence for His existence. If Plantinga were to establish the rationality of belief in God, clearly he would have to disable this argument. Plantinga has attempted to do so not by actually present- ing evidence for God’s existence, but by arguing that belief in God plausibly does not. need evidence to be rational. Conceivably, it may not be a nonfoundational belief, (which, as we have seen, both the foundationalist and the coherentist require evidence for), but rather a “basic belief" whose rational justification lies not in any inferential support it derives from other beliefs, but in the external circumstances which give rise to it. He argues that failure to recognize this possibility lies in the epistemic myopia of an "eviden- tialism" grounded in "classical foundationalism". Conse— quently, he presents arguments against this sort of eviden- tialism and in behalf of the possibility that Theism can simply be a basic belief, for if that is the case, then Theism can be rational even for one who has no evidence at all for it. In essence, he here follows the same philosophical tactics as before, attempting to clear the arena of epis— temological views which undermine Theism and then provide a rationale according to which Theism may well be rational for someone whose episi his argument were towards forever un objections to Thei of philosophical s 1.3 EVALUATING Ti If Theism is tion in the twenty epistemologically on pragmatic, his may at times nic. faith, but they c. appearances, the think they have g is to examine Plz 0f Theism in 0rde current threat C epistemic rectit' 22 someone whose epistemic condition is appropriate. Indeed, if his argument were to be successful, it would go a long way towards forever undermining one of the central philosophical objections to Theism. It is, therefore, an argument worthy of philosophical scrutiny. I-3 EEALflAIlHQ_IEE_BA§lQALlIX_QE_IflEl§M If Theism is to be a viable option for Western civiliza- tion in the twenty—first century, it will need to show itself epistemologically worthy of belief. Arguments for faith based on pragmatic, historic, or dogmatic concerns and assumptions may at times nicely supplement the rational foundations of faith, but they cannot support the entire structure. Despite appearances, the majority of people only believe what they think they have good reason to believe. The aim of this study is to examine Plantinga’s argument for the proper basicality of Theism in order to assess its adequacy as a response to the current threat of unbelief, for if he is correct, Theism’s epistemic rectitude has been greatly enhanced. This will require that we examine his negative attacks on classical foundationalism and coherentism as well as his positive proposal for the basicality of Theism. Since his argument for the possible proper' basicality of 'Theism is rooted in a broader epistemological discussion, I will concentrate (n1 expounding ins; epistemological position in general. The viability of belief in God as a properly basic belief obviously depends upon the Viability of the general epistemological fr in that larger coni central problem. epistemic and apol assuming the gene; position. The sec0nd . general. Followi his arguments aga tialism and brie basicality of bel The third ch basicality in or his POSition plat rests on its pro; able to Supply tl however, that t basicality which lend Theism the Moreover, I Will more SECure basi t0 Plantingals ( 23 epistemological framework in which it is placed. Deficiencies in that larger context will likely intrude themselves upon our central problem. We will also need. to reflect. upon the epistemic and apologetic value of so considering belief in God assuming the general validity of this wider epistemological position. The second chapter lays out Plantinga’s position in general. Following an early major article, it will expound his arguments against classical foundationalism and eviden— tialism and briefly present his opening defense of the basicality of belief in God. The third chapter analyzes Plantinga’s notion of proper basicality in order to see if it is adequate to the demands his position places upon it. If the rationality of Theism rests on its proper basicality, then such basicality must be able to supply the rationality Theism needs. I shall argue, however, that the weak foundationalist notion of proper basicality which Plantinga employs is not strong enough to lend Theism the epistemological foundation it requires. Moreover, I will contend that a rationale for epistemically more secure basic beliefs exists which does not fall victum to Plantinga’s objections. Foundationalism, I shall argue, needs stronger foundations than Plantinga’s position gives us, and so even if belief in God is "Plantinga-properly—basic", this will be of little help in restoring to Theism the sort Of epistemological rectitude it once possessed. Chapter IV argues that our belief in God does not gain the sort of ratic Plantinga basic be from deficiencies peculiarities of '1 But although basicality less tr his conviction th: factory epistemolw beliefs are nece cannot be accouni view of knowledge flawed. Thus, I foundationalism ; Theism nor an epi one is to assail '1 will need to empj The sixth ch warrant as it ap] compares them wi be shown that Wh; fall Victum to B( 24 the sort of rational support it requires even if it is a Plantinga basic belief. The reasons for this derive primarily from deficiencies in weak foundationalism, but epistemic peculiarities of Theism also contribute. But although I shall find Plantinga’s view of proper basicality less than satisfactory, the fifth chapter supports his conviction that classical foundationalism is an unsatis- factory epistemology. The main contention is that authority beliefs are necessary to our rationality, but since they cannot be accounted for by the "classical foundationalist" view of knowledge, classical foundationalism is irremediably flawed. Thus, I will agree with Plantinga that classical foundationalism is neither a necessary starting place for Theism nor an epistemological threat to its rationality. If one is to assail Theism for epistemological improprieties, one will need to employ other epistemic assumptions. The sixth chapter explores Plantinga’s views on epistemic warrant as it applies to both basic and nonbasic beliefs and compares them with BonJour’s modified coherentism. It will be shown that while on the one hand Plantinga's basic beliefs fall victum to BonJour’s demands for rational support, on the other hand, BonJour’s version of coherentism does not manage to escape Plantinga’s criticisms of coherentism. In fact, it will be shown that weak foundationalism and modified coheren— tism tend to converge with respect to what the former calls basic beliefs and the latter calls observation beliefs. The seventh chapter attempts to lay out a view of human an. m "a. :5. “surname! nun [Lucia 1 sipuoflfln :mc afifiquua fioihflflb #3311 9d: .yxoausiaijsz and? 2291 yiilsallld y is): =' A 7:-_:::f :--:7:x1':9:;i:‘.-.‘:.’)'J std ,seevherein such ism satisf ~2itsanu no ?i maiisnojvsh-Jn? I - -.-.'~-'..-.7.-:.- "'3': mi , After m '-.:;ri'torl:mr.= fresh} 1}. .-.:..- - -. -"-i' “ -"- c '5 - -:' ':.-.’ conclusion willsJ dammstrate the r 25 epistemology which accounts for "Plantinga properly basic be— liefs" without falling into either classical foundationalism, weak foundationalism, or coherentism. I will utilize some of Plantinga’s insights on the nature of epistemic warrant in order to develop a View of "rational noetic structures" which is more satisfactory than these alternatives so that we can ,see where in such a View belief in God should properly fall. After summarizing the findings of this study, the conclusion will suggest an alternative approach by which to demonstrate the rationality of Theism. TH As noted in Plantinga’s philO1 ality of religiou of different thei defense of belie traditional Weste "theistic belief' part, used by him God". As a whol that religious by dure, however, h rationality and demonstrate the v has involved two ranted belief CHAPTER II THE ATTACK ON FOUNDATIONALISM As noted in the introduction, the fundamental aim of Plantinga’s philosophical work has been to uphold the ration- ality of religious beliefs. While he has defended a number of different theistic beliefs, his main concern has been the defense of belief in God, i.e., belief that the God of traditional Western Theism exists. Indeed, "belief in God", "theistic belief", and "religious belief" are, for the most part, used by him as synonyms for "belief in the existence of God". As a whole, his program has been an endeavor to show that religious beliefs are rationally warranted. His proce- dure, however, has not been to develop a general notion of rationality and then apply it to religion, but rather to demonstrate the warrant of particular religious beliefs. This has involved two stages: First, showing that criteria of war- ranted belief which religious beliefs fail to meet are themselves either unwarranted.or at least highly questionable, and second, demonstrating that religious beliefs do meet those criteria for rationality which are valid (e.g., logical consistency). His more recent work on the epistemology of theistic belief has employed the same tactics: He first attacks what 26 he considers the in has been used ag construed by found rationality of the are now ruled out attack on evident: to formulate his I This will provide ing thereby the subsequent chapte tion will follow 60d" :1 (hereafter 11.1 m Plantinga's assault upon evi wone's belief 0u< of the evidence 1 27 he considers the most forceful epistemological position which has been used against theistic belief, "evidentialism" as construed by foundationalism, and then presents a case for the rationality of theism given that objections due to the former are now ruled out of court. This chapter will explore his attack on evidentialism and foundationalism, and then attempt to formulate his rationale for the warrant of theistic belief. This will provide a general overview of his position, furnish- ing thereby the context of the more specific analyses in subsequent chapters. For the most part, this general exposi- tion will follow his presentation in, "Reason and Belief in God"fi (hereafter, RBG). II.l EVIDENTIALISM Plantinga’s attack on foundationalism is part of a larger assault upon evidentialism, taken broadly as the View that "one’s belief ought always to be proportioned to the strength of the evidence for that belief." (RBG, 24) As expositors of this View, he cites Locke, Hume, Russell, and W. K. Clifford, quoting the latter as saying "’To sum up: it is wrong always and everywhere, and for anyone to believe anything on 1Alvin Plantinga, "Reason and Belief in God," Faith and Rationality, eds., Alvin Plantinga and Nicholas Wolterstorff (Notre Dame: University of Notre Dame Press, 1983), pp. 16— 93. See also his, "Is Belief in God Rational?", Rationality and Religious Belief, ed., C. Delaney (Notre Dame: University of Notre Dame Press, 1979), pp. 7-27; "The Reformed Objection to Natural Theology," Proceedings of the American Catholic Philosophical Association 54, 49-63; "Is Belief in God Properly Basic?", Nous 15, 41—51; "The Reformed Objection Revisted," Christian Scholars Review 11, 57-61. insufficient evidi words, holds not evidence in its su which do not have Plantinga notes, belief is to be 3 "’Give to any hypc ' ~ . just that rants. . . .1" (F More recent Scriven 90 even claiming that th. the burden of pr. claims that give existencyn (RBG is no evidenCe ac_ notes, ’God eXi aSYHImetricai, th latter requiring and 28 insufficient evidence.’" (RBG, 25) Evidentialism, in other words, holds not merely that warranted belief must have evidence in its support, but also that to believe propositions which do not have such backing is morally wrong. In addition, Plantinga notes, some expositors of evidentialism hold that belief is to be proportioned to evidence. Quoting Russell, "’Give to any hypothesis which is worth your while to consider . . . just that degree of credence which the evidence war- rants. . . .’" (RBG, 25) More recent expositors of evidentialism such as Flew and Scriven go even further, Plantinga continues, the former claiming that there is a presumption of atheism, i.e., that the burden of proof rests with the theist, while the latter claims that given the "’absence of any evidence for God’s existence'" (RBG, 27), atheism is obligatory, even if there is no evidence against God’s existence. As Plantinga quickly notes, ’God exists’ and ’God does not exist’ are, then, asymmetrical, the former requiring evidence for belief, the latter requiring only that there be insufficient evidence for the former. Finding Scriven's view too excessive, (RBG, 28-29), Plantinga formulates the essence of the evidentialist argument against God’s existence as follows: (1) It is irrational or unreasonable to accept theistic belief in the absence of sufficient evidence or reasons. and (2) There is no evidence or at any rate not sufficient evidence for the proposition that God exis Plantinga fir is going on in (2 proofs evidence (r He does not, how other difficulties one might also ad for rational belie doubtless would be accncept of evide which (a) the "em would be more war: latter would requ either met its 0. criteria for critw not be itself int 29 God exists. (RBG, 29) Plantinga first points out that a lot of question begging is going on in (2). Are not, he questions, the traditional proofs evidence (not proofs) of God's existence? (RBG, 29-30) He does not, however, pursue this difficulty. (There are other difficulties with (2) he does not mention. For example, one might also add that given the evidentialist requirement for rational belief, (2) itself needs evidence, evidence which doubtless would be hard to come by. In addition, (2) requires a concept of evidence and criteria for what counts as evidence which (a) the "evidence" for theism does not meet and (b) would be more warranted than any which theism did meet. The latter would require that one could show that such a concept either met its own criteria or most adequately met proper criteria for criteria. Furthermore, such a concept ought (c) not be itself incompatible with theism, for surely any re- quirement which by its very nature excluded theism would not of itself be a reasonable means of judging theism any more than one which entailed or presupposed it.) Plantinga’s main attention is directed at (1). He does not attempt so much to disprove it as to question its meaning and applicability. He agrees that indeed there are duties and obligations vis—a-vis belief. (RBG, 30) Think of those instances when a person’s life or well—being depends critical- ly upon the correctness of another’s beliefs and the conse- quent obligations the latter has to make sure his beliefs are correct. The surgeon, for example, who is about to slice someone open in 0: sure as possible of the problem and and (b) that he he Plantinga, hi nature of these di considered obligai where construing t seems wrong. He < 8 if your evaluat are higher than I taught by all th Plantinga notes, is no inference fl any search for e‘ failed to fulfill belieVes in God evidenCe of God, duties? Surely clear countEr-exa ed" obligation 1 (REG, 33) 30 someone open in order to remove his appendix ought to be as sure as possible (a) that the appendix is indeed the source of the problem and that removing it will solve the difficulty and (b) that he has the right person on the operating table.2 Plantinga, however, is principally concerned about the nature of these duties. Are they prima facie or all-things- considered obligations? (RBG, 33) He brings forth situations where construing them as an "all-things—considered" obligation seems wrong. He cites the example of a 14 year old (make him 8 if your evaluation and consequent expectations for a teen are higher than Plantinga’s) who believes in God because so taught by all the adults in his community. He does not, Plantinga notes, count their teachings as "evidence“; there is no inference from their belief to his. He believes without any search for evidence, but would we say, he asks, he has failed to fulfill his epistemic duty? That is, although he believes in God without having found or even looked for evidence of God, would we say he failed in his epistemic duties? Surely not, Plantinga holds, and so this seems a clear counter-example to there being an "all-things—consider- ed" obligation to believe only on the basis of evidence. (RBG, 33) Even the mature theist such as Aquinas who (wrongly let us assume) believes he has conclusive evidence for God’s 2There are, Plantinga notes, many ways to construct the ethics of belief-—teleologically, aretaically, and deontologi— cally. (RBG, 32) existence would r irrationally holdi in thinking he has violating (1), th after much dilige him of any real 9 construed as an al epistemic duty t1 evidence seems tc well as for chi apparently, be c, not have "suffic: Year old and Aqui the lack of evide Since it do: stemic duty as an then examines tr duty. 31 existence would be counted by few, Plantinga contends, as irrationally holding to theism. Although Aquinas may be wrong in thinking he has evidence adequate for rational belief, thus violating (l), the fact that he has reached his conclusion after much diligent and concentrated thought surely acquits him of any real guilt vis-a—vis epistemic duties. That is, construed as an all-things-considered obligation, the supposed epistemic duty to believe only on the basis of sufficient evidence seems to have clear counter—examples for adults as well as for children, for' Aquinas’ belief in God would, apparently, be considered rational despite the fact he does not have "sufficient evidence". (RBG, 33-34) Both the 14 year old and Aquinas, then, believe in God rationally despite the lack of evidential support for their belief. Since it does not seem plausible to construe this epi— stemic duty as an all-things-considered obligation, Plantinga then examines the alternative, taking it as a prima facie duty. Of course, as a prima facie duty, it can be rightly overridden in the appropriate circumstances. An emergency situation often makes normal evidence gathering either impos- sible or counter—productive, and in those circumstances, a person’s beliefs ought not be checked against the evidence as thoroughly as in less dire straits. Plantinga first presents as unconvincing an argument against (1) construed as a prima facie obligation which is based on the arguable assumption that beliefs are not under one’s direct control. Simply put, if I do not have control over my beliefs: refrain from a Paj Plantinga f because it seems hold people resp example of somec falsity of Nazism acts efficacious] holds, consider beliefs; surely t if he has done 5 guilt, surely, w actions he takes; those abhorrent would hold him g Upon them . "we 32 over my beliefs, I cannot have a prima facie obligation to refrain from a particular belief. Plantinga finds this response unconvincing, first, because it seems to him there are clear cases in which we do hold people responsible for their beliefsf He uses the example of someone who diligently questions the truth or falsity of Nazism, becomes convinced of its truth, and then acts efficaciously upon those beliefs. We do not, Plantinga holds, consider him culpable just because he acted on his beliefs; surely that is in one sense commendable, particularly if he has done so in the face of great difficulties. "His guilt, surely, does not consist solely in his taking the actions he takes; at least part of the guilt lies in accepting those abhorrent views." (RBG, 35) In fact, he argues, we would hold him guilty for those views even if he did not act upon them. "We do in fact sometimes think that a person is guilty--has violated norms or obligations--by virtue of the beliefs he holds." (RBG, 35) Here Plantinga’s rejection of the contention that unless we have control over our beliefs, lack of evidence places no obligation upon us to withhold them, seems to be based on the grounds that we do in fact hold people responsible for their beliefs. Even if such is the case, however, this response seems highly questionable. 3More accurately, in some of these cases, we hold persons responsible for holding certain beliefs; in others for not accepting certain beliefs. (RBG, 34) A second rea respond to an ev lability of belie without accepting distinction, one killing is wrong, to shoot the ene: believe another ‘ officer would nex the level of "fel by normal belief control. But ev accept) that bel t° 3“ “port it.‘ "belief" fOr "a Surely, regardle "subjectiVeu ser be the case. (F be had ceased be His mother; but true. In additio; directly Con tro 33 A second reason why Plantinga thinks the theist cannot respond to an evidentialist objector by citing uncontrol- lability of belief is because one can believe a proposition without accepting it and yige_y§;§a. To illustrate this distinction, one may not be able to help believing ‘that killing is wrong, but yet obey his commanding officer’s order to shoot the enemy because he chooses to accept as well as believe another belief of his, namely, that his commanding officer would never ask him to do what is in fact wrong. On the level of "felt belief", what one is "triggered to believe" by normal belief making mechanisms is perhaps beyond one’s control. But even if that is so, one can still reject (not accept) that belief on the cognitive level, thereby refusing to act upon it.“ Thus, the evidentialist need only exchange "belief" for "acceptance" to maintain his position, for surely, regardless of what we believe to be the case in the "subjective" sense of that word, we need not accept that to be the case. (RBG, 37—38) As Santyana has said, long after he had ceased believing in God, he still "believed" Mary was His mother; but, of course, he did not accept that belief as true. In addition, Plantinga notes that while we often cannot directly control our beliefs, we can control them indirectly, 4If this seems somewhat unclear, Plantinga recognizes the difficulty. He writes, "Now I am quite aware that I have not been able to make this distinction between acceptance and belief wholly clear. I think there is such a distinction in the neighborhood, however, and I believe it is important." (RBG, 38) and our obligati01 likely result in c measuring up to t‘ exhorted those c church, read the I eventually, to I calculus told the that if one canno its epistemic in Russell and Volta stay away from t] The real p evidentialist ob in God without St to belieVe Some requiring an in: 34 and our obligation can then be to do those things which will likely result in our eventually disbelieving propositions not measuring up to the evidentialist criterion. Much as Pascal exhorted those convinced by his wager argument to attend church, read the Bible, and pray in order to bring themselves, eventually, to really believe that. which the probability calculus told them was their best bet, so Plantinga suggests that if one cannot help believing in Theism despite realizing its epistemic unwarrantedness, he ought to read Bertrand Russell and Voltaire, attend rationalistic associations, and stay away from theistic authors. (RBG, 38) The real problem, Plantinga then notes, is why the evidentialist objector thinks one cannot rationally believe in God without sufficient evidence. Since it must be rational to believe some propositions without evidence on pain of requiring an infinite task of a finite mind, why does not belief in God qualify as one of them? (RBG, 39) This, of course, brings Plantinga to examine foundationalism, for it is here, he believes, that the argument for the "proper basicality" of belief in God must be established. II.2 FOUNDATIONALISM The culprit behind the evidentialist's insistence that belief in God be supported by evidence, Plantinga believes, is foundationalism. As an epistemological viewpoint, Plantinga recognizes that it has often been the assumption of evidentialist critics of theism, but one which many theists have adopted also Aquinas. While 1i Plantinga's eXPOS detail, it will be ments. What is impv Aquinas cle propositions ing, even th or reason to them on the I that a propo and do not be tions. (RBG Now the prop accepting as see to be t evident to PrOpositions ing them on] Ship to tho: nonbasic prc see to follc the Proposit US, do not 1 nonetheless Probable or AcIllinas meg PrOPOSition It is at lea are PrOperlx beliefs tha Senses. An tialist Obj Us only if . 35 have adopted also, the case par excellence being St. Thomas Aquinas. Plantinga’ While it is unnecessary for our purposes to follow 5 exposition of Aquinas' theory of knowledge in detail, it will be instructive to quote two concluding state- ments. What is important to see here is the following. Aquinas clearly believes that there are some propositions we are rationally justified in accept- ing, even though we do not have evidence for them, or reason to them from other propositions, or accept them on the basis of other propositions. Let us say that a proposition is basic for me if I believe it and do not believe it on the basis of other proposi- tions. (RBG, 46) Now the propositions we are rationally justified in accepting as basic, thinks Aquinas, are the ones we see to be true: those that are self-evident or evident to the senses. As for the rest of the propositions we believe, we are rational in accept— ing them only if they stand in a certain relation- ship to those that are properly basic. Among the nonbasic propositions we rationally accept, some we see to follow from those that are basic; these are the propositions we know. Others are not known to us, do not follow from basic propositions, but are nonetheless rationally acceptable because they are probable or likely with respect to them. I believe Aquinas means to hold, more generally, that a proposition is rationally acceptable for us only if it is at least probable with respect to beliefs that are properly basic for us——that is, with respect to beliefs that are self-evident or evident to the senses. And hence on his View, as on the eviden- tialist objector’s, belief in God is rational for us only if we have evidence for it. (RBG, 47)5 Given Plantinga’s construal of Aquinas’ foundationalism, then, we can posit four sorts of propositions. First, those that are self—evident or evident to the senses; second, those 5Plantinga notes, however, that there are hints of another View in Aquinas vis—a—vis God. "He sometimes suggests that there existence." is a sort of intuitive or immediate grasp of God’s (RBG, 47) that are deducible probable given th neither deducible The latter include those which are d and also those wh set. Although P1 to them, a fifth g -5 given set on intrinsic relati Since PrOpositio there would be a belief, then, is would be Confine be Considered u, tionalism' for agnosticism ratr for there Woul d 36 that are deducible from the first set; third, those which are probable given the first set; and finally, those which are neither deducible nor probable with regard to the first set. The latter includes those the probability of which is 0, i.e., those which are denials of propositions in sets one and two, and also those which are improbable with regard to the first set. Although Plantinga himself makes no explicit reference to them, a fifth group of propositions, those with probability .5 given set one——which of course includes those with no intrinsic relationship whatsoever to it--should be added. Since propositions are being construed as person relative, there would be an infinite number of the latter. Rational belief, then, is confined to groups 1-3 and irrational belief would be confined to group 4. The fifth set would doubtless be considered unworthy of belief for evidentialistic founda- tionalism, for there would be no evidence for them; but agnosticism rather than unbelief would be the proper attitude, for there would be no evidence for their denials either. A nonevidentialistic foundationalist, or at least a slightly attenuated one, could hold that such propositions are epi- stemic adiaphora and up to a person’s unfettered fancy to believe or not. Perhaps in such cases considerations other than evidentialist ones, say pragmatic ones, should be given a role in determining their belief—worthiness. Whatever, they certainly would be ruled out of court by those arguing that epistemic duty is dependent upon evidence. After expounding Aquinas’ View, Plantinga goes on to Plantinga as " . believes, together among him and the: analyzes the fo constituent parts integral to the ow basicalitynsome their relationsh are not belief~w< held more tentat Si°“"the extent particular belie believed-0n-the_ asymmetric and j strength to the foundational pr different fOUnc‘ 37 characterize foundationalism in general. Fundamental to his exposition is the concept of a "noetic structure", defined by Plantinga as “. . . the set of propositions . . . [one] believes, together with certain epistemic relations that hold among him and these propositions.“ (RBG, 48) Plantinga then analyzes the foundationalist noetic structure into its constituent parts and relations. Three ingredients which are integral to the construal of a proper noetic structure are (a) basicality--some beliefs are basic and it is in virtue of their relationship to basic propositions that others are or are not belief—worthy--(b) degree of belief--some beliefs are held more tentatively than others-—and (c) depth of ingres— sion-—the extent to which other beliefs are "tied into" a particular belief. (RBG, 48-50) The basic relation is "the believed-on-the—basis-of" relation, a relation which is both asymmetric and irreflexive. Belief is to be proportional in strength to the degree of "support" it receives from the foundational propositions. By "support", Plantinga notes, different foundationalists mean different things, but the general idea can be captured by construing it as meaning "evidence". That is, a proposition or set of propositions support a proposition or set of propositions just in case it (they) counts as evidence for the proposition or set of propositions. For each nonbasic proposition in the noetic structure, then, there will be a "path" from some basic proposition or set thereof to that proposition. The path is a set of propositions which counts as "evidence" for the nonbasic proposit: The key some as a basic proposi types of proposj formulations of f< tion “properly be which are eviden‘ versions only), a tic of modern ve1 first sentences : unmarried", “For its denial is fi mess", They in( and What We migl that "Self-evia. sorts of pr0p0sj "understand that 38 nonbasic proposition. (RBG, 52—57) The key concept, of course, is "basicality". What counts as a basic proposition and why? According to Plantinga, three types of propositions can be extracted from, the various formulations of foundationalism as qualifying for the designa- tion "properly basic", those which are self—evident, those which are evident to the senses (in classical and medieval versions only), and those which are incorrigible (characteris- tic of modern versions). Plantinga cites as examples of the first sentences such as "2+l=3", "No man is both married and unmarried", "For any proposition p the conjunction of p with its denial is false", and "Redness is distinct from green- ness". They include, in other words, logical, arithmetical and what we might call "broadly analytic" truths. He notes that "self-evident" means that simply by "grasping" these sorts of propositions intellectually one is able to "see" or "understand that" they are true. Some or much education may, of course, be necessary for one to be able to grasp properly the meanings of these propositions, but once they are rightly understood, their truth is evident. (RBG, 55-56) An important question, naturally, is what this "seeing" consists in. Plantinga notes that there are two components in the idea of self—evidence, one epistemic and one phenomeno— logical. "A proposition p is self-evident to a person S only if S has immediate knowledge of p--that is, knows p and does not know p on the basis of his knowledge of other proposi- tions." (RBG, 57) But in addition to this epistemic aspect, there is a "Phenm dationalists, he "luster" that "su: factor, that of accept these prop phenomenological they are held by self-evident trut Propositions “There is a tree 53) Plantinga Descartes, as opp hold that basic limit this categ see a tree". "incorrigible" , P is incorr POSSible t] it is not ; (RBG, 58) 39 there is a "phenomenological" component. According to foun- dationalists, he remarks, these propositions have a sort of "luster" that "surrounds" them.6 He also notes an additional factor, that of a strong, perhaps necessary, impulsion to accept these propositions. Perhaps, he ruminates, these two phenomenological aspects are one and the same, but regardless, they are held by foundationalists to accompany knowledge of self-evident truths. (RBG, 57) Propositions evident to the senses are exemplified by, "There is a tree before me", and, “I am wearing shoes". (RBG, 58) Plantinga notes that modern foundationalists such as Descartes, as opposed to ancient or medieval foundationalists, hold that basic propositions must be certain and, therefore, limit this category to those such as, "It seems to me that I see a tree". (RBG, 58) These are "immune from error", "incorrigible", and can be defined as follows: P is incorrigible for S if and only if (a) it is not possible that S believe p and p be false, and (b) it is not possible that S believe -p and p be true. (RBG, 58) Thus, propositions evident to the senses, accepted as properly basic by ancient and medieval foundationalists, are replaced by "incorrigible propositions" by modern foundationalists. It is at this point that Plantinga levels his strongest criticism at foundationalism. Foundationalism's "fundamental 6He refers here to Locke’s mention of an "'evident luster’" and the "’clarity and brightness to the attentive mind’" which he says they have, as well as to Descartes’ criteria of "’clarity and distinctness'". (RBG, 57) principle", accorl A propositio: and only if incorrigible (RBG, 59) It is, of cours Plantinga. Grant grounds can one r Plantinga ar the simple reasox for a Person Whit this noon", he w believe this pro Itake it as ba structure," (RE doing_ A fOUHdati respond that eit or that 1 am ir amore c°mpellil who accepts thw "self-referenti 40 principle", according to Plantinga, is the following: A proposition p is properly basic for a person S if and only if p is either self—evident to S or incorrigible for S or evident to the senses for S. (RBG, 59) It is, of course, the "only if" clause which troubles Plantinga. Granted that such propositions are basic, on what grounds can one hold that only these are properly basic? Plantinga argues first that such a thesis is mistaken for the simple reason that there are propositions properly basic for a person which do not meet this criterion. "I had lunch this noon", he writes, is such a proposition, for "I do not believe this proposition on the basis of other propositions; I take it as basic; it is in the foundations of my noetic structure." (RBG, 60) And, he adds, he is rational in so doing. A foundationalist, Plantinga notes, may, of course, respond that either I do not take such propositions as basic or that I am irrational in so doing. So Plantinga presents a more compelling argument, namely, that the foundationalist who accepts the given criterion for proper basicality is "self-referentially inconsistent" in so doing. That is, the criterion itself is neither self—evident, evident to the senses, nor incorrigible and, therefore, the foundationalist needs an argument from properly basic beliefs to the criter- ion. But alas, he claims, no such successful argument has as yet been given, so the foundationalist ought not accept this criterion. To insure th itwill be worth l mam. First, he ahst’s criterior (34) p is re (up 1 or 1nc< in S’s q1 . . self-e‘ rigibl Hethen continue: Now (34) h senses. Fur [the founda‘ it will be it is false If it is nor ly true or will be pos true; if th F believe 5 case is it not plausit such that o it. SO (3 senses, or therefore, (34)] he m tions tha- owident tc foundation argument f Condition. acceptance tion of re 41 To insure that no injustice is being done to Plantinga, it will be worth quoting his more formal version of the argu- ment. First, he gives another version of the foundation- alist's criterion of rational doxastic acceptability. (34) p is rationally acceptable for S only if either (1) p is self—evident or evident to the senses or incorrigible for S, or (2) there are paths in S’s noetic structure from p to propositions q1 . . . qn that (a) are basic for S, (b) are self—evident, evident to the senses, or incor- rigible for S, and (0) support p. He then continues, Now (34) itself is obviously not evident to the senses. Furthermore it is not incorrigible for F [the foundationalist]. If (34) is contingent, then it will be possible that F believe it even though it is false, in which case it is not incorrigible. If it is noncontingent, then it is either necessari— ly true or necessarily false. If the former, it will be possible that F believe it false when it is true; if the latter, then it will be possible that F believe it true when it is false; so in neither case is it incorrigible. Still further, (34) is not plausibly thought self-evident; surely it is not such.that one cannot understand it without believing it. So (34) is not self—evident, evident to the senses, or incorrigible for F. If (34) is true, therefore, then if F is to be rational in accepting (34), he must believe it on the basis of proposi— tions that are self-evident, incorrigible, or evident to the senses, and support it. But no foundationalist has ever produced a successful argument for (34) from propositions that meet that condition. It is therefore unlikely that F’s acceptance of (34) conforms to the necessary condi- tion of rationality (34) lays down. (RBG, 61) Plantinga then attempts one modification of (34), adding the clause, "is accepted as basic by nearly everyone,“ in order to justify accepting as basic propositions "entailing the existence of other persons and some propositions about the past," but ruling out the basicality of belief in God. (RBG, -c om criterion. foundationalist that belief in ' point, he holds overcomes his 0 properly basic, it sound. (RBG It should has neither re believing in Gor‘ for his disbeli nontheistic deti show why one wh she is irration needs support, 1 exists which 5} is to retain hr that the Theist Plantinga belie claim that all evident , or ev to be deemed r: 42 62) However, he argues, since many propositions he accepts as basic are not accepted by anyone else as basic (e.g., the above proposition about his having had lunch), this will not help. Moreover, any modification will face the same crucial objection as (34), namely, that it itself will not satisfy its own criterion. He adds that one could be a different sort of foundationalist or no foundationalist at all and still claim that belief in God needs evidentialist support. But at this point, he holds, such a claim needs an argument, one that overcomes his objections, that shows belief in God is not properly basic, and for which there is good reason to think it sound. (RBG, 63) It should be noted at this point that while Plantinga has neither refuted every epistemological reason for not believing in God nor laid the burden of proof on the atheist for his disbelief in God, nevertheless, he has placed the nontheistic detractor of theism in the position of needing to show why one who is a theist is irrational. The claim that she is irrational because she has no evidence for her belief needs support, Plantinga holds, because no convincing argument exists which shows she needs evidence for her belief if she is to retain her rationality. The evidentialist insistence that the Theist have evidence in support of her belief in God, Plantinga believes he has shown, rested on the foundationalist claim that all beliefs but those which are incorrigible, self— evident, or evident—to—the—senses require evidence in order to be deemed rational. But if the foundationalism upon which The final rationality of existence of ~- need any subst. . tions. His 5 : tradition begi in God was natu suitable condit we make an effo held we have ne Of course, eve holding that be the substantia‘ tific grounds v be tenuous, ev shown that we would not pro evolutionary 1 reasons which world, nevert under no obli 43 such evidentialism is based is itself questionable, then the claim that the Theist needs evidence for Theism (which is obviously nonbasic) if his belief is to be rational, is itself unwarranted. II.3 THE PROPER BASICALITY OF THEISM The final element in Plantinga’s argument for the rationality of Theism is his contention that belief in the existence of God is properly basic and, therefore, does not need any substantiation by argument from more basic proposi- tions. His standpoint is that of the Reformed theological tradition beginning with John Calvin. Calvin held that belief in God was natural, that we have a natural inclination under suitable conditions to believe in God and only disbelieve if we make an effort to overcome this inclination, much as Hume held we have natural inclinations for all sorts of beliefs. Of course, even if Calvin (and Plantinga) were right in holding that belief in God is natural and, therefore, basic, the substantiation of such a view on philosophical or scien— tific grounds would be difficult indeed, and the results would be tenuous, even if favorable. Moreover, even if it could be shown that we have such a natural inclination, that in itself would not prove God’s existence; there may well have been evolutionary reasons for the development of this inclination, reasons which, while based on survival value in the real world, nevertheless do not yield truth. But Plantinga feels under no obligation to argue for the claim that we do indeed .‘1' -' mandamus... .2‘ embrace. The ev he only needs t other objections is not a cons- development and certain obvious that if belief (2) that there all for holding beyond dispute, that he is defe 11.31 W Objection as basic, why I tions can one a belief to be basic for an)" and justified ogist" insist blanche propei 44 have a natural inclination to believe in God, but simply holds that belief in God can legitimately be basic for a person. That is, having rejected evidentialism and its foundationalist basis, he no longer needs arguments for belief in God, but can, for whatever reasons and in accord with a longstanding tradition, accept this belief as one which he can legitimately embrace. The evidentialist objection having been dealt with, he only needs to show that this claim does not succumb to other objections. Consequently, the remainder of his article is not a constructive argument for his position, but a development and elaboration of it in the context of answering certain obvious objections. The four he deals with are (1) that if belief in God is properly basic, any belief can be, (2) that there would, under this construal, be no ground at all for holding beliefs, (3) that belief in God would then be beyond dispute, a dogma rather than a rational belief, and (4) that he is defending a form of fideism. 11.31 The Great Pumpkin Objection Objection (1) is obvious. If belief in God can be taken as basic, why not any other belief imaginable? What restric- tions can one impose upon rational belief if one allows such a belief to be basic? Does not any belief become properly basic for anyone who has by any means he thinks appropriate and justified come to believe it? If the "Reformed epistemol- ogist" insists that he has not committed himself to carte blanche proper basicality, then what sort of criteria are to somesortof w'm And if the 1 characterize it it? Plantinga - First, the fac certain beliefs commits him to sort of belief who rejects a pa ion of meaning ‘ meaningful any does the Reform proper basical: he rejects the (RBG, 74—75) \ in God is prOP tion" to so III In certain air in God. Whel tendency is * under normal * 45 be used in disallowing some and not others? How can I be justified in claiming that Theism is, under certain condi- tions, properly basic, but Great Pumpkinism is not unless I have justifiable criteria by which to allOW' the one and disallow the other? Does not any such determination assume some sort of criteria, if not explicitly, at least implicitly. And if the latter, ought I not at a minimum be able to characterize it in some general terms or otherwise explicate it? Plantinga thinks not and gives 3 reasons for so thinking. First, the fact that the "Reformed epistemologist" holds certain beliefs to be basic in certain circumstances in no way commits him to the acceptance of the proper basicality of any sort of belief under any sort of circumstances. Just as one who rejects a particular formulation of the empiricist criter- ion of meaning does not thereby obligate herself to accept as meaningful any linguistic configuration whatsoever, neither does the Reformed epistemologist oblige himself to admit the proper basicality of just any proposition at all just because he rejects the foundationalist criterion of proper basicality. (RBG, 74*75) What Reformed epistemology holds is that belief in God is properly basic because there is a "natural disposi- tion" to so believe which has been built—in by the Creator. In certain circumstances, we humans have a tendency to believe in God. When one is in one of these circumstances, this tendency is triggered, just as when I stand before a tree under normal conditions, I have a natural tendency to believe which it is pre“ rejects the to does not mean th beliefs that st effect that the perly basic und tion, I'for any is properly has Second, P proper basicali judgments in re more than the 1 from making jut (RBG, 75) "Sup for the criter: nevertheless e that certain properly basic argues, altho propositions paradoxes), C propositions : 46 there is a tree before me. The fact that one holds that there are a number of circumstances under which belief in God is basic by no means obligates one to the View that "any belief is properly basic in any circumstances, or even to the vastly weaker claim that for any belief there are circumstances in which it is properly basic." (RBG, 74) Simply because one rejects the foundationalist's criterion of proper basicality does not mean that one is thereby forced to allow any and all beliefs that status. This, of course, is an argument to the effect that the proposition, "for person P, belief T is pro- perly basic under condition C", does not entail the proposi- tion, "for any belief S there exists condition C in which S is properly basic for some person P." Second, Plantinga holds that lack of a criterion for proper basicality does not itself prohibit one from making judgments in regard to the proper basicality of beliefs any more than the lack of a criterion of meaningfulness keeps one from making judgments about the meaningfulness of sentences. (RBG, 75) "Suppose I do not know of a satisfactory substitute for the criterion proposed by classical foundationalism; I am nevertheless entirely within my epistemic rights in holding that certain propositions in certain conditions are not properly basic." (RBG, 75) To put it another way, Plantinga argues, although some instances of seemingly self—evident propositions are not self—evident (witness the Russell paradoxes), one does not therefore reject all self—evident propositions; similarly, if it seems to you that you see a tree, "You would proposition that trees." Neithe withhold his ind belief in God ju: basic turned out not be rational : (RBG, 75) Lack does not, in otl or the rejectio basic but cannot valid criterion Plantinga’ there probably Proper basicali therefore, not justify the pro Great Pumpkinis typical fate of foundationalim 47 tree, "You would then be irrational in taking as basic the proposition that you do not see a tree or that there are no trees." Neither, then, must the Reformed epistemologist withhold his judgment in regard to the proper basicality of belief in God just because some beliefs which seemed properly basic turned out not to be so; for the same reasons, it would not be rational for him to hold the negation of belief in God. (RBG, 75) Lack of an explicit criterion of proper basicality does not, in other words, demand the withholding of judgment or the rejection of beliefs which you take to be properly basic but cannot prove to be so by virtue of some precise and valid criterion. Plantinga’s third argument consists in showing that (a) there probably are no universally justifiable criteria for proper basicality and (b) the Reformed epistemologist is, therefore, not responsible for putting forth one which would justify the proper basicality of belief in God and deny it to Great Pumpkinism. He argues for (a) by first pointing to the typical fate of proposed universal criteria for knowledge, the foundationalist criterion, for example. Since it does not meet its own conditions for basicness, it must be shown to follow from premises which do satisfy it if it is to be held rationally and in accord with foundationalism itself. But, he notes, no such argument exists and prospects for one do not seem promising. Note that it is its universality which makes it self—referential, and it is its inability to satisfy itself which lays it open to Plantinga’s charge of self—referential incoherence- If deduction from deduction seems same is true of basicality. It rigible, or (1‘ otherwise it wi fact is, I thini person S, A is incorrigible f revealing nece basicality fol clearly accept Given the sufficient or: reasons that a This would or conditions un 48 incoherence. If justifiable, then, it will have to be so by deduction from premises which do satisfy it, but such a deduction seems quite unlikely. Plantinga’s point is that the same is true of any proposed universal criterion of proper basicality. It will either have to be self—evident, incor- rigible, or deducible from propositions which are, for otherwise it will be open to dispute. Thus he writes, "The fact is, I think, that neither (2) [For any proposition A and person S, A is properly basic for S if and only if A is incorrigible for S or self-evident to 8] nor any other revealing necessary and sufficient condition for proper basicality follows from clearly self-evident premises by clearly acceptable arguments." (RBG, 76) Given the improbability of establishing a necessary and sufficient criterion for basicality by deductive means, he reasons that an inductive approach will more likely succeed. This would consist in taking examples of beliefs and the conditions under which they seem clearly to be basic and, correspondingly, conditions under which they seem clearly not to be basic, and then developing and testing hypotheses for proper basicality against these examples. Importantly, when this procedure is utilized, it becomes clear that the tradi- tional foundationalist’s criterion is wrong, for any number of beliefs which fail to meet that criterion are nevertheless rational for me to hold under certain conditions. For example, believing that what one ate for breakfast was as one remembers it later in the day, or that a human with whom I am conversing has t inductive fashi01 basicality ought and if a particu not the proper h: under suspicion. In other h arrive at a ur proper basicali‘ intuitions abop Plantinga admii conflicting lis intuitions, but this case the I to submit to cr for eXample) o justify-mg bel other beliefs basic_ His c pOSition abOU‘l 49 conversing has thoughts, feelings, etc. as I do.7 In good inductive fashion, Plantinga argues that criteria for proper basicality ought to conform to what we take as properly basic, and if a particular criterion does not, it is the criterion, not the proper basicality of the belief, that should be placed under suspicion. In other words, the fact that there seems no way to arrive at a universally' justified universal criterion of proper basicality means that we shall have to begin with our intuitions about what is and what is not properly basic. Plantinga admits that such a procedure will entail that conflicting lists will be composed by those with conflicting intuitions, but at this point his main contention is that in this case the Reformed epistemologist is not required either to submit to criteria proposed by others (the foundationalist, for example) or to come forth with a criterion which while justifying belief in God as properly basic, rules out all other beliefs which are properly not considered properly basic. His only obligation is to be able to justify his position about belief in God and Great Pumpkinism from the context of his own position. So, since Reformed theology holds there is a natural tendency to believe in God, but not to believe in the Great Pumpkin, Reformed epistemologists are perfectly within their epistemic rights to hold that the former is but the latter is not properly basic. 7See RBG, 76, small print, for a more precise explica- tion. '3? ‘{ E. w' 1'th 3‘1295' E": . '-: _|_', _ , Mus-uniti- Ina-nines“ m .mum- at: on ysilsoiued item on .aoiaiqsua tubal _'- L. ‘L' _'-'-,' .- I 50 war-.1.» m nun-r- m A 910M existfi (3) Edmund needi (4) An in taken (5) All s from (6) All s are t (7) Withm basi< used (3) A or a fr all‘ (9) A f: univ (10) What reje be i basi (11) The for: has: out is Plantf groundless. 50 Perhaps a schematization of this argument will be helpful. (1) A universal criterion for proper basicality must either justify itself, be deducible from propositions which satisfy it, or be establish- ed by an inductive process. (2) A self-justifying universal criterion for proper basicality or one deducible from propositions which satisfy it (likely) does not exists. (3) Therefore, an inductive procedure will (likely) need to be employed. (4) An inductive procedure will utilize beliefs taken as properly basic without criteria. (5) All such beliefs are considered properly basic from within a framework. (6) All such frameworks include assumptions which are taken as properly basic. (7) Without a universal criterion of proper basicality, no one set of assumptions can be used to show another set unjustified. (8) A criterion for proper basicality based upon a framework will (likely) not be accepted in all frameworks. (9) A framework need not, therefore, supply a universal criterion of proper basicality. (10) What a particular framework takes and what it rejects as properly basic need not, therefore, be justified by a universal criterion of proper basicality. (11) The Reformed epistemologist need not, there— fore, have a Indversal criterion of proper basicality which allows belief in God but rules out Great Pumpkinism. As Plantinga pointed out earlier, however, (RBG, 74), the Reformed epistemologist dOes not hold belief in God to be groundless. It is properly basic under certain conditions, and these conditi 1y basic, belief 11.32 will.“ In this sec basicality. AS cites, "1 an per ces there are cc belief, he point following sort justified in t “unusual" or "a not suffice (e. cinatory drug) belief is pro; then form the (RBG, 80) So "groundless" c or "conditions 51 and these conditions constitute its grounds. So while proper- ly basic, belief in God is not, for Plantinga, groundless. II.32 The Ground of Belief in God In this section, Plantinga refines the notion of proper basicality. As an example of a justified basic belief, he cites, "I am perceiving a tree". Since in appropriate instan- ces there are conditions or circumstances which "ground" this belief, he points out that in all such cases there will be the following sort of true proposition: "In condition C, S is justified 311 taking 1) as basic". (RBG, 79) If certain "unusual" or "abnormal" conditions exist, such a ground will not suffice (e.g., if one is high on LSD or some other hallu- cinatory drug). But Plantinga’s "central point" is that "a belief is properly basic only in certain conditions" which then form the justification for or "ground of" this belief. (RBG, 80) So, he reasons, properly basic beliefs are not "groundless" or "gratuitous", but have justifying conditions or "conditions that confer justification on one who accepts them as basic." (RBG, 82) II.33 Is Argument Irrelevant to Basic Belief in God? In this section, Plantinga makes it clear that proper basicality does not place a belief beyond argument. Defeat- ers, defeater—defeaters, etc. are all possible and demand response if a belief is to retain its proper basicality, for proper basicality confers prima not ultima facie justification upon proposition: 11.34 Meier It might bi belief in God 5] because faith he basic. This is he claims that i arguing for the likens this bel Trinity or Inca: in the truthfulr but to beliefs Perceptual trut these "natural believed with01 Prepositions f1 rivhtly so bel for they form lustified by m Starting POim 52 upon propositions. II.34 giggigm It might be though that Plantinga is telling us that belief in God should be accepted "on faith" and that it is because faith has been invoked that this belief is properly basic. This is a false interpretation, however, for in fact he claims that belief in God is a deliverance of reason. In arguing for the proper basicality of belief in God, Plantinga likens this belief not to those such as the doctrine of the Trinity or Incarnation, beliefs which must be taken "on faith" in the truthfulness of some purported revelation or authority, but to beliefs like self—evident truths, memory truths, and perceptual truths. It is not faith, but reason which finds these "naturally" belief—worthy and basic, for they are believed without evidence (at least evidence in the form of propositions from which I infer them), but are nevertheless rightly so believed under certain (and most) circumstances, for they form the set of those beliefs which (a) cannot be justified by more fundamental ones and (b) themselves are the starting points of justificatory chains in evidentialist arguments for nonbasic beliefs. Plantinga is not arguing that religious faith should be added to those belief forming dispositions by which some of our natural beliefs are formed, but is claiming that our natural and rational belief forming dispositions produce, properly in certain conditions, belief in God. Therefore, under the proper conditions, belief in God as a properly be of reason, “Ct f II.4 CONCLUSIO From this at least three First, it accept "evident in God as such propriety of supplies certaJ' God does not : further points not reject the of all religic within religiv utilize and r between Calvix references to 53 as a properly basic belief is, for Plantinga, a deliverance of reason, not faith. (RBG, 90) II.4 QQHQLHfilQfl From this preliminary summary of Plantinga’s position, at least three general conclusions can be reached. First, it is apparent ‘that. while Plantinga. does not accept "evidentialism" as applicable to theism, ie., belief in God as such does not need evidence, he does not reject the propriety of evidence or the rationality which evidence supplies certain beliefs. He is only claiming that belief in God does not require evidence for its rationality. Three further points need to be made in this regard. First, he does not reject the use of or need for evidence for the rationality of all religious beliefs. Presumably, arguments between and within religious traditions and belief systems would both utilize and require varying sorts of evidence. Arguments between Calvinists and Arminians, for example, would require references to biblical texts which supported a proponent’s views and conflicted with those of his adversaries. Hindus and Muslims might argue over the evidence for reincarnation. Christians and Muslims might debate the veracity of the Bible and the Koran in order to decide the facticity of the cruci- fixion of Christ, the doctrine of the Trinity, or the Incarna— tion. In all these cases the mustering of evidence would be necessary, and there is no reason to believe any of the propo— sitions involved would ever be properly basic for Plantinga. He is only COD exists". Second rejecting the ne not advocating believe in Thei various sorts of edge acquisitio of evidence, P1 in turn that on ces as part of foundationalis‘ means of logice these experien means of" or ' experiences ca 1), Plantinga Theism. He wo if present, n faith if they they are unne Second, ti<>nalism as f0m Which on to the Senses aetually are First, he is He doubts no. 54 He is only concerned with the proper basicality of "God exists". Second, even with regard to "God exists" he is only rejecting the need for foundationalist type evidence. He is not advocating that one is justified in simply deciding to believe in Theism. One comes to such a belief by means of various sorts of experiences, encounters, training, and knowl- edge acquisition, and if one insisted that these were a type of evidence, Plantinga would not demur. He would only insist in turn that one did not use these various sorts of experien— ces as part of a foundationalist noetic structure, i.e., as foundationalist evidence. One need not arrive at Theism by means of logical inferences from the propositions that express these experiences, but might simply believe "through" or "by means of" or "on account of" them; or perhaps better, these experiences can validly "give rise to" belief in God. Final- ly, Plantinga does not say that evidence is irrelevant to Theism. He would not reject foundationalist sorts of evidence if present, nor would he feel they would somehow undermine faith if they were forthcoming. He is merely holding that they are unnecessary for rational faith. Second, Plantinga has not, in effect, rejected founda— tionalism as such, but only a strict form of it, namely that forniwhich only accepts self-evident, incorrigible, or evident to the senses propositions as properly basic. In fact, he is actually arguing for two modifications of this criterion. First, he is arguing that the "only if" clause be stricken. He doubts not that self-evident, incorrigible, and evident to evidence if our and, clearly, t? unless the critl can reasonably it from properl ties seem impo would seem bes Second, he is tions need to among these is that it is pro certain persm Presents argum Should be so 501119 theologi; When it meets 55 the senses propositions ought to be accepted as properly basic in appropriate circumstances, but rejects the claim, plausib- ly, that these are the only sorts of propositions that should be so accepted. Doubtless he is correct in holding that a number of other propositions need to be accepted without evidence if our noetic structure is to get off the ground; and, clearly, there is no way to retain the "only if" clause unless the criterion can be formulated in such a way that one can reasonably claim proper basicality for it or can argue to it from properly basic beliefs. But both of these possibili- ties seem impossible, and given the present situation, it would seem best to simply leave the criterion open ended. Second, he is claiming not only that more sorts of proposi— tions need to be included as properly basic, but also that among these is "God exists". Of course, he is not claiming that it is properly basic under any conditions, but only for certain persons in certain situations. For this claim he presents argumentation, so again, he is not asserting that it should be so taken by everyone or just on the authority of some theologians, but merely that it can be properly basic when it meets the required stipulations. Third, one should not see Plantinga’s position as a rejection of ea generally empiricist epistemology. Taking belief in God as properly basic does not entail an epistemol- ogy of divine illumination or rational insight. In fact, for two reasons his claim can be seen as a broadening and deepen— ing of empiricist epistemology. First, by holding that more sorts of PrOPOSj to experience i' In essence, he what is not pro One must examin and then see if an "inductive" it is clear th; Plantinga, invc proper basicali circumstances . very appeal to empiricism. N1 a belief depen rise to it, th properly basic ical) facts" 0 to assume that revisable. p] the proper bas behalf I argumh tion, he Clea about the pro in regard to revisable to heartily Welc 56 sorts of propositions can be properly basic, he is appealing to experience itself as the criterion of proper basicality. In essence, he is saying that one cannot decide what is and what is not properly basic by doing "arm chair" philosophy. One must examine the various sorts of propositions there are and then see if and when they might be properly basic. Such an "inductive" procedure is typical of empiricism. Second, it is clear that any appeal to proper basicality will, for Plantinga, involve an appeal to experience. The case for the proper basicality of a proposition will rest on the sort of circumstances in which belief in it arises, and it is this very appeal to experience which, again, is characteristic of empiricism. Not only will the possible proper basicality of a belief depend on the sort of experiences which might give rise to it, the propriety of a (specific case of) belief as properly basic will depend on an investigation of "the (empir- ical) facts" of the situation. Moreover, there is no reason to assume that assessments of proper basicality will not be revisable. Plantinga is not presenting an a priori case for the proper basicality of Theism, but giving arguments in its behalf, arguments which are themselves defeasible. In addi- tion, he clearly allows for the revisability of judgments about the proper basicality of a belief, both in general and in regard to particular cases. Hence, epistemology becomes revisable to its foundations. But such a view should be heartily welcomed by empiricists. ed “WM . I ' 11W. :1: Quinn au- _ 3, . - ' 5 - ‘6 3m ”“1"” 9 ha ha, m--um “shimmer an .. ‘ USU? .3135” 541.30,..‘16‘ _:_,. .' *nzlin; Yin-fl: _n-_-:'.".'a: s:-".r . _' v.33 and: u our Mani . '.l ’ ' v- f) D". - 'I!" "-i3:ubfii“ “a l [EM-Ifi'fifgg - . _. :_'.'- - '_ '-_ .. ‘1. . '5' i-_ _ ' -_ ‘ basicality 0f 1 - '51.}, r , ' :' l general is a 13! consists in bd basic beliefs beliefs which 1 IIclassical for proper basical. for basic bel necessary for of basic belie: Not only is tn limit oneself CHAPTER III PROPER PROPERLY BASIC BELIEFS Our initial survey of Plantinga’s argument for the proper basicality of belief in God shows that his epistemology in general is a form of modest foundationalism. This modesty consists in both less stringent requirements for properly basic beliefs and the consequent broader range of basic beliefs which this leniency allows. But while rejecting the "classical foundationalist" (hereafter CF) criterion for proper basicality, Plantinga does not at all reject the need for basic beliefs. They are, he believes, epistemically necessary for justified belief; it is simply the narrow range of basic beliefs to which CF limits one that must be modified. Not only is there no demonstrated epistemic obligation to so limit oneself, such a limitation, he feels, also unduly eliminates a great number of epistemically warranted rational beliefs which neither have nor need foundationalist evidence, to say nothing of the innumerable rational beliefs based on them. Although "Reason and Belief in God" might give the impression that Plantinga wants to make belief in God basic and as a result has framed his general epistemology to suit his purpose, such a conclusion would be quite mistaken. Even if belief in God were never basic for anyone, in his view CF 57 a would still be c But what 50 and what is the first section 0 of basicality, I tion of basic b cussion of his criterion for w the appropriate The second seci account of pro essentially ar basic beliefs i (loals. The 1; minimal criter: use it to arguv itv is a bit more radical foundationalis foundations. III. . 1 m In a four 58 would still be a faulty and indefensible epistemology. But what sorts of beliefs do count as basic for Plantinga and what is the justification for accepting them as such? The first section of this chapter (III.1) will examine his view of basicality, laying out his understanding of the justifica- tion of basic beliefs as basic. It will then move to a dis— cussion of his View of warrant, for it turns out that the criterion for warranted belief in general is what determines the appropriate grounds for accepting some beliefs as basic. The second section (III.2) will discuss problems which this account of proper basicality presents for foundationalism, essentially arguing that Plantinga's construal of properly basic beliefs is inconsistent with foundationalism’s epistemic goals. The last part (III.3) will attempt to construct a minimal criterion of proper basicality for foundationalism and use it to argue that Plantinga’s construal of proper basical- ity is a bit too moderate. Properly basic beliefs must be more radical in their truth claims if they are to provide foundationalist noetic structures with sufficiently firm foundations. III.1 PLANTINGA OE PROPER BASICALITY In a foundationalist epistemology, properly basic beliefs are those beliefs on which all other beliefs ultimately depend for their justification. Foundationalists have traditionally thought, therefore, that these beliefs must somehow be "self— justifying" and/or certain. The sort of foundationalism Plantinga finds criteria for pr Plantinga beliei which do not me observation bel their status as them as basic? beliefs to be t Plantinga “Reason and Bel not needing evi Observation be] "of a certain (here, for exa crucial role" justification. °f appearance lustifies tak Pearance is 59 Plantinga finds defensible, however, does not accept such criteria for proper basicality. Indeed, as we have seen, Plantinga believes we can accept as basic a number of beliefs which do not meet these criteria, e.g., memory beliefs and observation beliefs. But what, then, grants these beliefs their status as basic or justifies the believer in accepting them as basic? After all, we don’t want to allow just any old beliefs to be taken as basic. Plantinga gives a general answer to this question in, "Reason and Belief in God." Basic beliefs, he argues, while not needing evidentialist support, are nevertheless grounded. Observation beliefs, for example, are "grounded" in experience "of a certain sort". (RBG, 79) The appropriate experience (here, for example, being appeared to in some way) "plays a crucial role" in both the formation of the belief and its justification. (RBG, 79) The experience of a certain sort of appearance, in other words, both causes the belief and justifies taking it basically.1 Since the role of the ap- pearance is in both cases "crucial", we can assume that without it that observation belief either would not have arisen or would not be justified as basic.2 In such a 1Later in the paragraph, Plantinga adds, "We could say, if we wish, that this experience is what justifies me in holding it; this is the ground of my justification, and, by extension, the ground of the belief itself." (RBG, 79) So while basic beliefs do not have evidential justification, they do have justification, yiz., the doxastic conditions which justify believing a particular belief basically. ’With the proviso, of course, that no other epistemically warranting cause of that belief was operating at that time. person at a tin and is within h (b) his noetic then accepting 5 circumstances) in Chapter II, 1 each sort of pr: sition such as as basic." (R? propositions as a particular bi denoted by "R" So basic belie: which occur 60 situation, what we might call a "justified in believing basically" situation, ". . . a belief is justified for a person at a time if (a) he is violating no epistemic duties and is within his epistemic rights in accepting it then and (b) his noetic structure is not defective by virtue of his then accepting it." (RBG, 79) The experience (in appropriate circumstances) "justifies" or grounds the belief. As noted in Chapter II, Plantinga points out that this implies that for each sort of properly basic belief there will be a true propo— sition such as, "In condition Q, S is justified in taking 9 as basic." (RBG, 79) (From here on, I shall refer to such propositions as "rules" which designate under what conditions a particular belief type is properly basic, and they will be denoted by "R". The set of all Rs will be denoted by {R}.) So basic beliefs are grounded in experiences of a certain sort which occur in the appropriate contexts. As Plantinga concludes, The central point here . . . is that a belief is properly basic only in certain conditions; these conditions are, we might say, the ground of its justification and, by extension, the ground of the belief itself. In this sense basic beliefs are not, or are not necessarily, groundless beliefs. (RBG, 80) Plantinga applies this view to belief in God. Certain conditions, he holds, call forth belief in God, conditions such as ". . . guilt, gratitude, danger, a sense of God’s presence, a sense that he speaks, perception of various parts of the universe." (RBG, 80) Given these sorts of circumstan— ces, beliefs such as the following are properly basic: "God is speaking to 1 proves of what 3 to be thanked a1 given this seen is not strictl} those just list however, remain such as, "There “I see a tree. beliefs entail speaking is pei One must l the justificat believer’s bel of a proposit: above guotatic are “the grou thrust of Plan Of belief do 61 is speaking to me," "God has created all this," "God disap- proves of what I have done," "God forgives me," and "God is to be thanked and praised." (RBG, 80) He then concedes that given this scenario, the general proposition, "God exists," is not strictly speaking basic; only propositions such as those just listed are. The parallel with perceptual beliefs, however, remains. What is basic is not general propositions such as, "There are trees," but more specific ones such as, "I see a tree." (RBG, 81) But given that such specific beliefs entail the more general beliefs, this "loose" way of speaking is permissible. (RBG, 82) One must be careful not to confuse two separate issues, the justification of the belief and the justification of the believer’s believing. The former concerns the justification of a proposition, and despite Plantinga’s statement in the above quotation to the effect that the specified conditions are "the ground of its [the belief’s] justification," the thrust of Plantinga’s position seems to be that the conditions of belief do not (directly) justify the belief itself. The belief is basic and, therefore, need not be justified as such. What the conditions do is to "confer justification on one who accepts them as basic." (RBG, 82) That is, the conditions do not justify the proposition believed, but rather justify the believing posture of the one who under those circumstances believes the proposition. Contrast a belief based on evidence with a properly basic belief. Belief p is justified by evidence e. What confers jUStification or has. 50 3'5 be] which justifies would not be jus sort of justifi justified in be justifies p. Contrariwi basically if S does not need belief for hi1 evidence, or b: know that p* i him. If he di in order to kn some R which 5 that the cond: one would nee. made R PrOperI c0nditions wh is justified if it is to b n“ my do without that BY grounding which they w Satisfy that 62 justification on S's belief that p is the evidence e which S has. So S's believing is justified by his knowledge of that which justifies p, namely, e. Without a knowledge of e, S would not be justified in believing p (if, of course, no other sort of justification for p were available to S). So S is justified in believing p only if S knows whatever it is that justifies p. Contrariwise, however, S is justified in believing p* basically if S believes p* under the right conditions. He does not need to know whatever it is that makes p* a basic belief for him, whether it is p*'s incorrigibility, self— evidence, or basicality under C. That is, S does not need to know that p* is basic for him in order for it to be basic for him. If he did, nothing would ever be basic for anyone, for in order to know that p* is basic, one would need to know both some R which states the conditions under which p* is basic and that the conditions designated by R hold. But to know this, one would need either evidence that R holds or some R* which made R properly basic for him, and so on ad infinitum. So the conditions which make p* basic for S do so whether or not S is justified in believing that p* is basic. Foundationalism, if it is to be true to its raison d'etre, needs beliefs which not only do not require evidence, but also can be believed without that believing requiring evidential justification. By grounding basic beliefs in conditions or circumstances in Which they would be properly basic, Plantinga has sought to satisfy that need. The view 0 then, can be 51111 tional beliefs, belief -worthY - (perhaps more a by the conditic set of conditii So the bottom c basic beliefs basic is provii the conditions the members of ed, but would 1 is any develo Properly basi< R to be basi. conditions ar as bilSic give membership in like any 0the upon the bel it. Clearly “0t have the for in basic they cannot Certainty, 1 63 The View of foundationalism which Plantinga presents, then, can be summarized as follows. First, there are founda- tional beliefs, beliefs which do not need evidence to be belief—worthy. The belief-worthiness of these beliefs (perhaps more accurately, the believing of them) is granted by the conditions C in which they are had. Under a certain set of conditions, certain beliefs are legitimately basic. So the bottom of our noetic structure consists of the set of basic beliefs {B} and the rationality of accepting these as basic is provided by the set of propositions {R} which state the conditions under which each member of {B} is basic. How the members of {R} are determined as such is left undevelop— ed, but would proceed by means of an "inductive" process. Nor is any development given to the reasons for thinking that properly basic beliefs need conditions as specified by some R to be basic, nor to the difficulty of determining when conditions are such that it is proper to accept some belief as basic given certain conditions, i.e., the criterion for membership in {B}. The rest of a noetic structure will look like any other foundationalist structure, each belief relying upon the belief(s) below to confer evidential support upon it. Clearly, in Plantinga’s View, properly basic beliefs do not have the sort of indefeasibility one traditionally looked for in basic beliefs. They are basic not in the sense that they cannot be wrong or that they supply certainty or maximal certainty, but rather in the sense that they are rational beliefs which dC support. Warra internal proper' self-evidence, such as indefee property of th! Rather, it is t which supply wz neither of bel stances. 0f c those circumst tion to other are to be warr stances in toi But just relation" beti tion? What, epistemic war basicality? eDistemic st; Criterion tn. c0nditions . 64 beliefs which do not obtain their rationality from evidential support. Warrant for these beliefs does not come from any internal property which they have such as incorrigibility or self—evidence, nor from satisfying some epistemic criterion such as indefeasibility or maximal certainty, nor from some property of the agent such as "doing one’s epistemic best". Rather, it is the circumstances which give rise to the belief which supply warrant. In other words, warrant is a property neither of beliefs nor of persons, but of believing circum- stances. Of course, the belief and the believer are part of those circumstances, and clearly they must be in proper rela— tion to other aspects of the total situation if the beliefs are to be warranted. But warrant is a property of the circum- stances in toto. But just what all is involved in there being a "proper relation" between the believer and his total epistemic situa— tion? What, in other' words, does a full explication of epistemic warrant entail and how does it all relate to proper basicality? There is much more to "warrant" or "positive epistemic status" for Plantinga than the purely externalist criterion that beliefs are basic if produced in the proper conditions. In "Justification and Theism",3 Plantinga introduces the criterion of one’s epistemic faculties "working properly" as one of the necessary conditions for warrant. Clearly, such 3Alvin Plantinga, "Justification and Theism," Faith and Philosophy, IV (1987), 403—26. a criterion als‘ Plantinga write: 1 . . positive e equipment, ing appara It must be to functic us. Intit positive I that my producing ties are w way they V Of course necessary for working propel that addition of warrant. C us than others more by way 01 memory belief 390 I owned a both these b 65 a criterion also assumes some sort of "design plan". Plantinga writes, I . . . suggest that a necessary condition of necessary for warrant. positive epistemic status is that one’s cognitive equipment, one’s belief forming and belief sustain- ing apparatus, be free of cognitive malfunction. It must be functioning in the way it was designed to function by the being who designed and created us. Intitially, then, let us say that a belief has positive epistemic status, for me, to the degree that my faculties are functioning properly in producing and sustaining that belief; and my facul— ties are working properly if they are working in the way they were designed to work by God.4 Thus, Of course, more than just "working as designed" is working properly in an appropriate environment.5 But One’s epistemic equipment must be even that addition does not give us a sufficient characterization of warrant. Clearly, some propositions have more warran us t t for han others. Surely, Plantinga insists, "Modus ponens has more by way of positive epistemic status for me than does the memory belief, now rather dim and indistinct, that forty years ago I owned a second hand 16 guage shotgun. . . J” Although both these beliefs, he argues, have been produced by the proper functioning of his epistemic equipment in the appro— priate environment, the former is clearly more warranted than the latter. . . a belief B has positive epistemic status for S if and only if that belief is produced in S by his epistemic faculties working properly (in an ap- propriate environment), and B has more positive ‘Ibid., p. 407. 5I bid., pp. 408—9. ‘Ibid., p. 409. Consequently, his final definition is as follows: epistemic 1 tive epist' not or elSj B than 8*- Clearly, t rant are proper corollary of a epistemic syst« conditions for is the result c osition specif proper functic to what degree system forms t In "Pro; Plantinga dev \ 7M”: Function", 1; 1M, James 66 epistemic status than B* for S iff (l) B has posi— tive epistemic status for S and (2) either B* does not or else S is more strongly inclined to believe B than 8*.7 Clearly, the key elements in Plantinga’s concept of war- rant are proper functioning in appropriate conditions and the corollary of a "design plan" which determines both how our epistemic systems are to function and what the appropriate conditions for such functioning are. In other words, each R is the result of epistemic design, for each such rule or prop- osition specifies what the appropriate conditions are for the proper functioning of one of our belief forming systems and to what degree we are justified in believing the beliefs that system forms under those conditions. In "Proper Epistemic Status and Proper Function",a Plantinga develops the concept of warrant from a different 7Ibid., p. 410. In "Positive Epistemic Status and Proper Function", in Wow 1988, James E. Tomberlin, ed. (Atascadero, California: Ridgeview, 1988) (hereafter PESPF.), after giving the same definition, Plantinga continues, "If we think degree of belief and degree of positive epistemic status are real valued functions, we can put the matter thus: where a person S accepts a proposition P, S has positive epistemic status to degree r for P if and only if his faculties are functioning properly in producing this belief and S accepts P to degree r*, where r* is the value for r under the appropriate func- tional transformation." (PESPF, 34) _ He also specifies the following: (a) Proper functioning is belief specific. That is, one’s epistemic faculties must be operating properly in regard to the production and sustain- ing of the specific belief in question. (b) Proper function- ing and proper environment are not absolute. Approximations to the ideal are permissible. (c) Malfunctioning epistemic apparatus and malappropriate environments can be compensated for if I know the former are not functioning properly or the latter are not properly appropriate. (PESPF, 36-37) 8See fn. 7. direction, wt 1 characterizes " tive property t epistemizes trl that the three "Chisholmian 11 not guarantee e which a persor satisfies the these positior warrant for B. of these posit is "cognitive faculties to f has positive apparatus is to work in pr Planting POSitive epis this conceptj beliefs form appropriate e truth , u and 67 direction, but reaches the same conclusion. He first roughly characterizes "positive epistemic status" as ". . . a norma- tive property that comes in degrees, enough of which is what epistemizes true belief." (PESPF, 3) He proceeds to show that the three currently popular views on epistemic warrant, "Chisholmian Internalism," coherentism, and reliabilism, do not guarantee epistemic warrant. Cases can be constructed in which a person holds some belief B in such a way that he satisfies the criterion for epistemic warrant which each of these positions dictates, and yet fails to have epistemic warrant for B. The common denominator in the failure of each of these positions to adequately account for epistemic warrant is "co nitive malfunction, failure of the relevant cognitive faculties to function properly." (PESPF, 32) Thus, "A belief has positive epistemic status for me only if my cognitive apparatus is functioning properly, working the way it ought to work in producing and sustaining it."9 (PESPF, 32) Plantinga then goes on to develop the same definition of positive epistemic status as quoted above. He recognizes that this conception of epistemic warrant assumes (a) ". . . that beliefs formed by our faculties functioning properly in an appropriate environment will be for the most part close to the truth," and (b) a "design plan" which determines to what gAs Plantinga is quick to point out, this does not mean functioning normally "in the statistical sense." Many normal functions of our cognitive apparatus give rise to false beliefs, e.g., day dreaming or wishful thinking. Thus the "ought" in the definition. degree and in W Although P late a Gettier does think tha‘ of Gettier prob is the producti failure of th (PESPF, 43) I] given the pro; apparatus for paratus to deg B to degree r2 gives rise tc these stipula be constructi construal of situation is internally, t lent to sayij anY Way. th¢ slll‘prising, \ 68 degree and in what circumstances (a) holds.lo (PESPF, 39) Although Plantinga states that his aim is not to formu— late a Gettier-proof analysis of knowledge (PESPF, 40), he does think that this perspective plainly reveals the source of Gettier problems. "What is essential to Gettier situations is the production of a true belief despite a relatively minor failure of the cognitive situation to 'match its design." (PESPF, 43) In other words, for any belief B and believer S, given the proper functioning to degree r of S’s epistemic apparatus for B in an environment appropriate to that ap- paratus to degree r1 for B, S will be warranted in believing B to degree r2.11 Moreover, if Plantinga is correct about what gives rise to Gettier problems, one would think that when these stipulations are met, no Gettier counter—example will be constructible. This may well be correct, for what this construal of epistemic warrant says is that if my noetic situation is not in any way defective, either externally or internally, then my belief is justified. But this is equiva— lent to saying that if my justification is not defective in any way, then my belief is justified. True, but hardly surprising. What Plantinga has contributed to other such lo"What confers positive epistemic status is one’s cognitive faculties working properly, or working according to the design plan insofar as that segment of the design plan is aimed at producing true beliefs." (PESPF, 39. Emphasis his.) He adds that the "design plan" may incorporate change with time, including change in the design plan itself, and that design plans are species specific. (PESPF, 39-40) 1&3 will be a function of r and rv attempts to res epigone is the be in proper WC which they are have justified operating unde in which it w:‘ perhaps the 10 recent episten be such that a epistemic envi plan for the 1 design plan i fair number necessary and \ 12There Gettier exam] upon either a SiIStem. Whex because I 1533 there is not functioning ( 69 attempts to resolve the difficulties posed by Gettier and his epigone is the proviso that one’s own noetic faculties must be in proper working order and suited to the conditions under which they are working. This, of course, is true. One cannot have justified beliefs if one’s belief forming mechanism is operating under conditions for which it was not designed and in which it will likely not lead to true beliefs. Indeed, perhaps the long sought "4th requirement", the Holy Grail of recent epistemology, is simply that the cognitive situation be such that a proper match exist between the cognizer and his epistemic environment, i.e., one which accords with the design plan for the former. It may even be, in addition, that this design plan is discernable at least to the extent that in a fair number of situations we can evaluate whether or not necessary and sufficient conditions for knowledge are met.12 12There are, however, two difficulties. First, many Gettier examples do not seem to rely in any particular way upon either a faulty noetic environment or a faulty epistemic system. When I believe that someone in my class owns a Ford because I falsely but justifiably believe Smith owns a Ford, there is nothing in either my epistemic environment or the functioning of my cognitive faculties which is malappropriate or malfunctioning. Second, if one construes "malappropriate— ness" and "malfunctioning" broadly enough so that they cover all Gettier situations, including mine vis-a—vis Smith, my class, and Ford ownership, then of course Gettier problems cannot arise. But that is merely to say that everything involved in justifying a belief must be such that the belief is truly justified if the belief is to be justified. While true, that is hardly revealing. If I assume that my epistemic mechanisms, when working properly in the proper circumstances, will always give me truly justified beliefs, then I can rule out all counter-examples by claiming a malfunction or malap— propriateness of some sort. But since every sort of error producing possibility is thereby included in these categories, it would seem so doing does not help at all in determining what precise qualifications must be met by a particular belief for it to be warranted. What, in other words, is the exact Regardless relates to prO] tions which mL' basic is dete conditions are problems aside beliefs which relationship c case if my bel cannot tell i appropriate er. I judge whetht merely know th tio“; I must malfunction is Ford ownershi Smith and Ford between my epj in this situa malfunctionint I could, propriateness. PM way of be infinitely the Potential cOmpleted the be.f°rehand wh with a partic I cannot tell cOTI'Strual of 70 Regardless, the thrust of Plantinga’s position as it relates to properly basic beliefs is that the sort of condi- tions which must be satisfied for a belief to be properly basic is determined by the design plan, and when those conditions are met, that belief is properly basic. Gettier problems aside, we at least have a set of properly basic beliefs which form the foundation of our noetic structure. relationship of my evidence to my belief which must be the case if my belief is to be justified? Unless I know that, I cannot tell if my faculties are functioning rightly in an appropriate environment, and only if I know that they are can I judge whether or not my belief is warranted. I must not merely know that there is some malappropriateness or malfunc— tion, I must know what the exact malappropriateness or malfunction is. Just what is wrong with my deduction about Ford ownership in my class from my justified belief about Smith and Ford ownership? In what way is there a misalignment between my epistemic faculties and my epistemic circumstances in this situation? Or just how is my epistemic equipment malfunctioning here? I could, of course, begin compiling a list of malap- propriatenesses and malfunctionings, but (a) there is no a priori way of knowing how long such a list would be; it might be infinitely long; (b) there is no way to determine what all the potential problems on the list are until one has actually completed the list; and (c) I would have no way of knowing beforehand which of the possible problems on the list is wrong with a particular belief. Without this knowledge, however, I cannot tell if my belief is justified. So while Plantinga’s construal of warrant may indeed tell us what conditions are necessary for a belief to be warranted, and while it may have given us sufficient conditions for warranted belief, it will not help us determine whether or not a particular belief is warranted. Without a completed list of potential problems which is known to be complete, we cannot be sure our belief does not somehow succumb to one. In other words, if our justified beliefs must meet the requirements Plantinga sets out for warrant, we must know that a particular belief satisfies them for that belief to be warranted for us. Of course, the opposite track might be attempted. I could list all the appropriate circumstances and proper functionings, so that if my particular case were not on that list, I would know my justification was inadequate for rational belief. The same problems, mutatis mutandis, would nevertheless remain. 111.2 mum Plantinga modest f em. to accept witl tions. These of the chem "grounded" by epistemic con following two First, 1 distinction t and evidence to ground my count as evi particular e Point to e t sidered rati My believing the other ha 71 III.2 EBQELEM§_flllfl_ELAflIlE§A_BBQBEBLX_BA§L;JEEQEE§ Plantinga’s foundationalism, it can be seen, is of a very modest form. Beliefs are basic as long as they are rational to accept without evidence under a particular set of condi- tions. These basic beliefs possess this basicality because of the circumstances which give rise to them and so are "grounded" by these circumstances. They are grounded in epistemic conditions rather than epistemic evidence. The following two difficulties can be raised against this View. First, from an internalist perspective, can a valid distinction be made between conditions which ground a belief and evidence for a belief? If certain circumstances are said to ground my belief b, do not those same circumstances also count as evidence for the belief—worthiness of b? If a particular evidential belief be is called into question, I point to e to substantiate it and that belief is then con— sidered rational because I am in possession of that evidence. My believing be is justified because In is justified. If, on the other hand, the basicality of a basic belief b of mine is called into question, I would, given Plantinga’s View, recount the conditions under which it takes place. I must show that it is basic and, therefore, rational by showing that it arose under the proper epistemic conditions. But does not the recounting of these circumstances count, then, as evidence that b is a basic belief for me? Moreover, does not part of my case for the basicality of that belief involve my using some R (some rule stating that under conditic] a case that R i that b is basi< the truth of b the circumstax count as reas< for the truth In order for the basic that belief i‘ First, 0 Consider an a intuit as ane I do not un limited intel the meaning c {M (the set given reliab indirectly 3‘) true because 72 under conditions C belief b is basic)? So I also have to make a case that R holds as well. In that case, does not showing that b is basic also count as warrant for b itself, i.e., for the truth of b? For how can believing some b be justified if the circumstances which justify that believing do not also count as reasons for believing b true? And are not reasons for the truth of b what we mean by evidence for b? In order to see how the evidence contained in an argument for the basicality of a belief can be indirect evidence for that belief itself, let us examine 3 similar cases. First, compare the justification of an analytic belief. Consider an analytic truth a which for some reason I cannot intuit as analytic. For example, it might be in a language I do not understand, or its complexity' may transcend. my limited intellect, or it may employ words which I do not know the meaning of. If, however, I know that all analytic beliefs {A} (the set of all analytic truths) are true, and then am given reliable evidence that a is a member of {A}, a has been indirectly justified for me by virtue of that evidence. a is true because it is a member of {A}, but it is known by me to be true because of the evidence I have that a is an element of {A}. For me, that evidence is evidence for a itself. Consequently, my belief that a is true is based on the evidence I have that a is a member of {A}. It is, therefore, neither self-evident nor basic for me, even though it might be for someone else. Likewise, if a basic belief b is known by 'me to belong to {B} (the set of basic beliefs) and, therefore, like the evidence f itself. Similarly counts as evid of that "evide truth of the t is an epistemi theorem. But else, since I dence" that i broad constrt belief for me Finally scripts with Suppose I ha' reconstructi 73 therefore, likely true because it is a member of {B}, for me, the evidence for membership in {B} is also evidence for b itself. Similarly, compare a mathematical theorem t. Its proof counts as evidence that it is a theorem, and it is by virtue of that "evidence" that I know that theorem, even though the truth of the theorem does not depend on the proof. The proof is an epistemic tool by which I gain permission to believe the theorem. But although that theorem may be basic for someone else, since I believe it on the basis of the deductive "evi- dence" that it is a theorem (a member of {T)), it is, in this broad construal of evidentialism, a derivative or evidenced belief for me. Finally, take an empirical proposition such as, "Manu- scripts with letters of style 2 date from 810 to 870 A. D." Suppose I have a severely marred manuscript but that by some reconstructive techniques can manage to make a case, despite the illegibility of the letters, that they were of style 2. Then by giving evidence that some object is a member of a certain class of objects, I have come to know that it has certain properties which belong to all members of that class, despite the fact that I could not have learned that members of that class of objects have those properties directly from that item itself. That is, garnering indirect evidence that a proposition p belongs to some set of true or likely true propositions provides evidential support for p as much as does direct evidence in its behalf, and in a doxastic situation which relies ‘ evidential bel Therefore and empirical some belief b then no less supported by calling the 1 virtue of de "directu evid So evide means of ind: less evident rational beca that, I must just as I d belief , say \ ”By "di SUCh as I e ' ( that not alI sense of "di 74 which relies (mi such indirect evidence, I) is likewise an evidential belief, one justified by evidence. Therefore, it would seem that for analytic, mathematical, and empirical truths occasions arise in which evidence that some belief b is a member of {B} is evidence for b, and b is then no less supported by evidence than were it directly supported by some evidence e. Let us distinguish these by calling the former "indirect evidence" (i.e., evidence by virtue of demonstrating' class membership) and the latter "direct13 evidence" for p. So evidence that a belief is belief-worthy can be by means of indirect as well as direct evidence, and it is no less evidential for that reason. Thus, if a belief is rational because it is properly basic, then in order to show that, I must give direct evidence for its proper basicality, just as It do when II mount direct evidence for some other belief, say the date of St. Augustine’s birth.14 But then I 13By "direct" we mean evidence for the proposition itself, such as, e.g., the exhibiting of a white raven as evidence that not all ravens are black. Of course, evidence in this sense of "direct" can be indirect in the sense that deductions or other intervening propositions have to be employed for the evidence to actually count as evidence. For example, evidence that Sam Spade was in jail during a bank heist is, in this latter sense, "direct" evidence that he was in jail and "indirect" evidence that he did not partake in the robbery. In the sense used in the text, however, it is also "direct" evidence that Spade is innocent of the robbery. 1“Note that for any evidential belief be there is a rule R_which designates the conditions under which beis warranted, so evidential beliefs are also warranted because they arise in accord with some rule. Moreover, the rules stating the conditions under which evidential beliefs are warranted will differ from those stating the conditions under which basic beliefs are warranted only insofar as the former contain lists am giving indi thereby making Note, in properly baSiC I must give d some R15 and, S particular cas I clearly givr So what 6 between wide the demonstr: therefore, be can see, for evidence, ant There a1 the distinct justificatio basically. \ 0f approprie ”That : Warrants th ev'ldence tn; the Conditi 75 am giving indirect evidence for that (basic) belief itself, thereby making it an evidential belief. Note, in fact, that in order to show that a belief is properly basic, I must give evidence on two levels. First, I must give direct evidence that b is properly basic under some R15 and, second, I must give direct evidence that in this particular case b satisfies R appropriately. But by so doing, I clearly give indirect evidence for b itself. So what epistemically significant distinctions are there between evidence given in support of an evidential belief and the demonstration that. a particular' belief is basic and, therefore, belief-worthy without evidence? None as far as I can see, for in the final analysis both are supported by evidence, and, therefore, so is that basic belief itself}6 There are two obvious replies. First, one can point to the distinction made above between evidence for a belief and justification for the rationality of believing something basically. R sets out the conditions of rationally believing of appropriate evidence. 15That is, there must be some rule Re (see fn. 14) which warrants the inclusion of R within {R}, so there must be evidence that R is a member of {R}. But since R just states the conditions under which b is basic, the evidence showing that R satisfies 1%, and, therefore, is a member of {R}, indirectly shows that b is basic; that is, it shows that b is a member of {B}. 16In other words, if basic beliefs are grounded in certain conditions, then there is evidence in their behalf, and while someone may not in fact believe them on the basis of that evidence, if he is to defend them or his belief of them, he must point to that evidence. b basicallY (0] this distinct plantinga’s Vi given the prop! in turn, is "<5 belief b can under conditi< I am justifie< fications for be shown to s b is basic, 1 that b is bas itself seem t the conditim b is likely ‘ Second, not evidence be held to b true. Plar beliefs, not that a bel 76 b basically (or for b's basicality), not evidence for b. But this distinction cannot be maintained. According to Plantinga's view, {R} would be the set of all R which hold given the proper functioning of our epistemic equipment which, in turn, is "designed" to result in true beliefs. So if any belief b can be shown to be basic, it is likely to be true under conditions designated in the appropriate R. Therefore, I am justified in believing b basically if b meets the quali- fications for basicality, yi;., R. But then insofar as b can be shown to satisfy R, that counts not only as evidence that b is basic, but also as evidence for b, for what R says is that b is basic and, therefore, likely true. Indeed, R would itself seem to be evidence for b since a belief is basic under the conditions delineated by R only if it can be shown that b is likely true given R. Second, one might hold that evidence that b is basic is not evidence for the truth or likely truth of b. A belief can be held to be basic without thereby holding that it is likely true. Plantinga is arguing for the basicality of these beliefs, not their truth. But what sense does it make to say that a belief can be basic for someone without thereby implying that in those conditions in which it is properly basic, it is likely true? Would not such an admission as this reply contemplates not only be a severe departure from foundationalism, but also a serious undermining of the very rationale for having basic beliefs to begin with? Perhaps basic beliefs do not need to be indefeasible, but clearly they must be likely those beliefs structure wil? arriving at tr it is not com Plantinga rejecting all strictly ext Indeed, our e an externalis on basic beli a problem whi that Plantim include intei A secor beliefs is t] their truth be procured. basic, but 77 must be likely to be true, for if not, neither will any of those beliefs based upon them, which is to say that a noetic structure will then function for some purpose other than arriving at the truth. This may be the case, but certainly it is not compatible with Plantinga’s program. Plantinga could still defend his position by simply rejecting all internalist concerns, i.e., by maintaining a strictly externalist position vis-a-vis basic beliefs. Indeed, our exposition above makes it clear that his view is an externalist one. But then the viability of his position on basic beliefs depends upon the feasibility of externalism, a problem which shall be addressed later. Indeed, I will show that Plantinga’s own criticisms of externalism lead him to include internalist elements in his construal of warrant. A second difficulty with Plantinga’s View of basic beliefs is that they are defeasible, and that entails that if their truth needs to be established, evidence for them must be procured. But then, it would seem, they are not properly basic, but justified by the evidence provided by other beliefs, gig, that evidence which would properly be mustered in response to whatever evidence could be used to defeat them. We have just seen that there is a formal need for evidence to justify these basic beliefs as basic. They are established as properly basic by virtue of satisfying some R. But that can only be the case if in fact there exists evidence both that R holds (surely R cannot be properly basic) and that a particular belief does satisfies some R iJI a specific situation (the: fulfilled for properly basil conflicts witl Then the evide direct" natur evidence for t a class of pr must then def proper basica belief itself at all. However, beliefs whict even if there I am rationa evidence; th< them, am I r have reason 1 17 As tI-o SuCh aw: ltself- Tm Shows t at Stifficient Supply be a 78 situation (there must be some evidence that conditions C are fulfilled for b). So now suppose one of these defeasible properly basic beliefs is challenged by evidence which conflicts with it or by other justified sources of doubt. Then the evidence required is no longer of a formal and "in— direct" nature; it must be concrete and "direct", i.e, evidence for b itself and not just evidence that b belongs to a class of propositions likely true under conditions C.17 I must then defend that belief not merely by pointing to its proper basicality, but by mounting support directly for the belief itself, and then it will no longer be properly basic at all. However, properly basic beliefs are to function as beliefs which need no evidence, which are rational to hold even if there is no evidence in their behalf. It may be that I am rational in accepting a whole ream of beliefs without evidence; the question is, if no evidence in fact exists for them, am I rational in continuing to believe them if I also have reason to doubt them? Surely I cannot just fall back on their proper basicality to substantiate their truth, for proper basicality does not insure truth. I can only look to the evidence that does exist for their truth, but, again, then they cease to be properly basic for me. 17As we have argued, evidence that a proposition belongs to such a class is indeed also evidence for that proposition itself. Thus, a defeating claim must include evidence which shows that this class membership is not, in this instance, sufficient to warrant the proposition. Otherwise, it would simply be a contrary claim and not a potential defeater. That prop may not be pro‘ person at a Plantinga’s V? beliefs one wi noetic struct giarantee the in fact the e‘ not one feel them as "pro1 be in order foundational in question? Properly bas tions for c falsity of f course, for memory belil breakfast, ; won1d doubt What F is rationa \ “"Pla: belief" am Properly b basic thos @0625 and < 1ncluded, 79 That properly basic beliefs for one person at one time may not be properly basic for another person or for that same person at another time is certainly compatible with Plantinga’s view. But then they hardly seem the sorts of beliefs one would want at the base of their foundationalist noetic structure, for if they are of this sort, they only guarantee the truth of those beliefs which depend on them if in fact the evidence in their behalf is sufficient. So ought not one feel obliged to check that evidence before accepting them as "properly basic"? Would not such scruples at least be in order when a "Plantinga properly basic belief"18 of foundational importance for one's entire noetic structure was in question? If so, one would not then be believing it as properly basic. In particular, given its immense ramifica- tions for one’s noetic structure, would not the truth or falsity of Theism call for such epistemic scrupulosity? Of course, for matters of everyday life, such as relying on memory beliefs in order to tell your friend what you had for breakfast, an attempt to support these Plantinga basic beliefs would doubtless be overly scrupulous. What Plantinga’s claim seems to boil down to is that one is rational in accepting certain beliefs without evidence. 18"Plantinga properly basic beliefs", "Plantinga basic belief" and variations thereof will be used to refer to those beliefs which he accepts as basic but. which are not CF properly basic. Of course, Plantinga would accept as properly basic those beliefs which are CF basic too, but for the most part, I will use this designation in order to denote those he does and CF does not so accept. When CF basic beliefs are included, that will be obvious. must be suppOI call them "pl limitations 01 to accept 501 beliefs ought ga’s view, th evidential be their behalf their case ne correct here basic“ in am or Theism. I I am permitte until some 0. But the: foundational Foundational beliefs, and 80 The declaration that they are properly basic establishes nothing helpful, for they are defeasible and when challenged must be supported by evidence. But if this is the case, why call them "properly basic"? Why not say that, given the limitations of human epistemic powers, it is rational for us to accept some beliefs without evidence while some other beliefs ought never to be accepted in this way. In Plantin- ga’s View, the difference between properly basic beliefs and evidential beliefs is not that the former have no evidence in their behalf and the latter do, but that evidence is not in their case necessary if conditions are right. Even if he is correct here, that in itself does not make them "properly basic" in any sense that does much good for foundationalism or Theism. It simply means that given a particular situation, I am permitted to hold that belief without evidence unless or until some cause for significant doubt arises. But then these beliefs cannot be basic in the sense that foundationalism, classic or otherwise, needs basic beliefs. Foundationalism needs basic beliefs that will ground all other beliefs, and that without further evidence. But Plantinga’s basic beliefs will not do that. They themselves, when challenged, must often be grounded in evidence, and in fact even if never challenged are themselves grounded in the evidence for their proper basicality under the conditions specified by some R. Plantinga properly basic beliefs, in other words, do not seem to supply what foundationalism needs in its properly properly basic beliefs. Here one migl these objections ti replies to these thesis. The clai' therefore, cannot times it is ratio: their basicality evidence for them stature as basic : tions specified bj evident belief m: example, does not who believe it be Still self-evide: though one can gi‘ a self-evident p Proper basicality likewise, the fac basic beliefs Car these basic beli. does not take awa is being defendm beli“ basically either the belief ity is not affec 81 Here one might defend Plantinga by claiming that both these objections to his view and the criticisms of the initial replies to these objections seriously misunderstand his thesis. The claim is not that some beliefs are basic and, therefore, cannot be supported by evidence, but only that at times it is rational to believe them without evidence. That their basicality can be evidentially supported and that evidence for them can be presented does not diminish their stature as basic for those who believe them under the condi- tions specified by the appropriate R. The fact that a self— evident belief might also be supportable by evidence, for example, does not take away from its self—evidence for those who believe it because it is self—evident. For them it is still self-evident and, therefore, basic. Moreover, even though one can give "indirect" evidence for the basicality of a self-evident proposition, namely its self-evidence, its proper basicality is not thereby diminished or erased. So likewise, the fact that the basicality of Plantinga properly basic beliefs can be supported by evidence and the fact that these basic beliefs themselves can be supported by evidence does not take away at all from their proper basicality. What is being defended is the rationality of one who takes such a belief basically even if he is unaware of any evidence for either the belief itself or its basicality, and that rational- ity is not affected one way or the other by the presence of evidence of any kind relevant to the truth of that belief. In short, Plantinga’s thesis is externalist, while these objections all pr) I shall retu: my point is not 1 their justificati Plantinga properl muster. I am clz believed without dational beliefs evidence, direct the contrary, la of CF evidence a rational beliefs Even se, t1 miqht he were properly basic k to those belief then their bas-l evidence for . hOWQVer’ Signi beliefs and P1; ferences Show 1 the epistemic 82 objections all presuppose internalism. I shall return to this issue in later chapters, but here my point is not that justified beliefs require knowledge of their justification in order to be justified, but simply that Plantinga properly basic beliefs are not up to foundationalist muster. I am claiming that even if they can legitimately be believed without evidence, they are not satisfactory foun— dational beliefs; not that unless believed on the basis of evidence, direct or indirect, they are not rational. Quite the contrary, later it will be argued that despite their lack of CF evidence and their nonbasicality, they are nevertheless rational beliefs. Even so, this externalist response is not adequate. It might be were Plantinga properly basic beliefs, like CF properly basic beliefs, basic for some reason intrinsic either to those beliefs themselves or to the believing of them, for then their basicality would not need to be grounded at all in evidence for them or for their basicality. There are, however, significant differences between CF properly basic beliefs and Plantinga properly basic beliefs, and these dif- ferences show that Plantinga basicality is not sufficient for the epistemic purposes of foundationalism. First, CF properly basic beliefs are no wit more belief— worthy if external evidence in their behalf exists or not, for that belief-worthiness resides in the nature of the belief itself. Plantinga properly basic beliefs, however, do become more belief-worthy if "external" evidence exists in their behalf. It may b tree when appeare strengthened by C God is rational w: conditions, but 5 also have a pm belief that I am 2 further evidence I can only cite ‘ Second, CF certain, approe properly basic many, perhaps ex bility is of su only the most t: be supported by would seem the belief-worthy t classical found hold they sing Properly basic Thirdly, foundational I 83 behalf. It may be that I am justified in believing I see a tree when appeared to treely, but that belief is certainly strengthened by confirming evidence. Perhaps even belief in God is rational without any supporting evidence under certain conditions, but surely that belief too is strengthened if I also have a gmoof for God’s existence. Contrariwise, my belief that I am appeared-to-treely is not strengthened by any further evidence, and if challenged, as Chisholm emphasizes, I can only cite the belief itself. Second, CF properly basic beliefs, if not absolutely certain, approach certainty asymptotically. Plantinga properly basic beliefs, however, are no more certain than many, perhaps even most, evidential beliefs. Their defeasi— bility is of such a fragile nature that they are "basic" in only the most tenuous circumstances. Indeed, since they can be supported by external evidence and often need to be, it would seem that evidential beliefs are quite often more belief-worthy than Plantinga properly basic beliefs. So while classical foundationalist basic beliefs are among (some would hold they simply are) our most certain beliefs, Plantinga properly basic beliefs are not necessarily among these at all. Thirdly, CF properly basic beliefs are up to being foundational, for they are the sort of beliefs which can ground other beliefs but which themselves can receive no firmer grounding than they already have as basic beliefs. Plantinga properly basic beliefs, however, do not seem up to the task of being foundational, for they themselves (a) are not the court of evidence. They 5‘ rather than the r‘ sought. One might beliefs can funct structure only if Thus, perhap of basic beliefs basicality. The basic beliefs i believing experii ditions which gi to support such 1 which they deriv will require evi properly basic (i.e., if that beyond the need be some quality ing experience Section will an a general Crite It may well be, 84 not the court of last resort and (b) are often in need of evidence. They seem the proverbial foundation built on sand rather than the rock which foundationalism has traditionally sought. One might even conclude that Plantinga properly basic beliefs can function meaningfully in a foundationalist noetic structure only if they can be grounded in CF basic beliefs. Thus, perhaps the main problenlwith Plantinga’s construal of basic beliefs concerns the locus of the grounds for their basicality. The rationale for accepting Plantinga properly basic beliefs is external to both those beliefs and the believing experience itself. It resides, rather, in the con- ditions which give rise to those beliefs. But then in order to support such beliefs, the bridge between them and that from which they derive will need to be made, and that, obviously, will require evidence. In other words, if a belief is to be properly basic in the sense demanded by foundationalism, (i.e., if that belief is to be properly properly basic and beyond the need of supporting evidence), there will need to be some quality or property of either the belief, the believ— ing experience or both which grounds the belief. The next section will attempt to justify this claim and to establish a general criterion of proper basicality for foundationalism. It may well be, of course, that there are no such beliefs, but if not, then we need to know that and forsake once and for all the temptation foundationalism offers us. If, however, foun— dationalism is to be a viable epistemology, its basic beliefs must be able to satisfy the criterion developed in the next section. III.3 319w Plantinga’s ‘ denies not only evident, but also opens up the posi thought to requ actually be props not arbitrary 01 whatever he wisi particular belie such conditions admit to this e memory beliefs, are few types 0: tions under whi main problem in is basic would belief, but in ( belief is had eVidence, We ] strual of prop foundationalisn fail t0 Provid distinction bet 85 section. III-3 BBQEEB_BBQEEBLX_§A§lQ_BELlEE§ Plantinga's View of properly basic beliefs, we have seen, denies not only that they need 'be incorrigible or self- evident, but also freely admits to their defeasibility. This opens up the possibility that a great many beliefs formerly thought to require evidence for rational acceptance may actually be properly basic. That basicality, to be sure, is not arbitrary or ungrounded; nor can one will to believe whatever he wishes. Certain conditions must exist if a particular belief is to be rationally accepted as basic, but such conditions are construed broadly enough by Plantinga to admit to this esteemed society beliefs based on authority, memory beliefs, and observation beliefs. Most likely, there are few types of beliefs that would have no possible condi- tions under which they could be accepted as basic, and the main problem in determining whether or not a particular belief is basic would not be deciding upon its classification as a belief, but in determining if the conditions under which that belief is had are such as to warrant that belief without evidence. We have given reasons to think that such a con- strual of proper basicality is inadequate to the needs of foundationalism, for not only do such properly basic beliefs fail to provide a sufficiently secure foundation, the very distinction between foundational and nonfoundational beliefs seems at the least blurred. I propose, therefore, to present an alternative. What sort of proper basicality the reasonablenes not do so by some test would show empirical test co such beliefs as t to? Epistemic o by current practj bility exists the direct intuition claimed, for one that intuition. have to be formr Plantinga I dure in Order tc belief type is 1 how Such a proc hard to envisic all if We have belief Ought tc basic, Moreov reviSable’ Wil Proper basical The Only analysis of w‘. 86 an alternative. What sort of an argument does a prospective criterion for proper basicality need? How, in other words, can we establish the reasonableness of accepting a criterion? Surely we could not do so by some empirical method, for what sort of empirical test would Show us that such a criterion was correct. An empirical test could reveal whether or not most people do take such beliefs as basic, but how could it show that they ought to? Epistemic oughts, like moral oughts, cannot be decided by current practice, for the distinct, perhaps likely, possi- bility exists that current practice is faulty. Doubtless also direct intuition of a proposed criterion’s truth cannot be claimed, for one could reasonably ask for a substantiation of that intuition. So to justify such a claim, an argument will have to be formulated in some other fashion. Plantinga has suggested that we use an inductive proce— dure in order to determine whether or not a certain belief or belief type is legitimately taken as properly basic. Exactly how such a procedure would proceed is not developed and is hard to envision, but certainly it cannot be carried out at all if we have no idea what sort of thing a properly basic belief ought to be if it is to be properly taken as properly basic. Moreover, such an initial idea, although doubtless revisable, will at least act as an intuitive criterion for proper basicality. The only method adequate to such a predicament is an analysis of what proper basicality means and how properly basic beliefs are structure. Thiss proper basicality beliefs must have tion for a noetic necessary to 91'0“ Plantinga’s defe qualification, b criterion can be course, contain ‘ to be basic, for est version of f< cide ahead of t That could, perk Foundation: 0f certain know Will be more or their relations thrust of found is to assure tj them in the sux 87 basic beliefs are to function in a foundationalist noetic structure. This section, then, will argue for a criterion of proper basicality on the basis of the nature properly basic beliefs must have if they are to provide the sort of founda- tion for a noetic structure which, foundationalism argues, is necessary to ground empirical knowledge. We have argued that Plantinga’s defeasible basic beliefs do not meet such a qualification, but it remains to be seen if an adequate criterion can be constructed. Such a criterion should, of course, contain the minimum standards necessary for a belief to be basic, for a foundationalist need not defend the strong- est version of foundationalism. Nor need such a criterion de- cide ahead of time which beliefs qualify and which do not. That could, perhaps, be discovered "inductively". Foundationalism does not, of course, hold out the promise of certain knowledge for dependent beliefs. Indeed, these will be more or less probable given their evidential base and their relationship to properly basic beliefs. However, the thrust of foundationalism, its primary purpose if you will, is to assure the belief—worthiness of beliefs by grounding them in the surest basis possible. Only if our beliefs are grounded in beliefs which need no further argumentation or evidence, the foundationalist claims, will our beliefs have any warrant at all. His basic case for this position is twofold. First, all other ways of grounding beliefs are invalid (coherentist views, e.g., are circular) and, there- fore, lead to scepticism. Second, if all beliefs are justified by belie have an infinite cannot carry out a is argued, amount of a criterion f foundational adec further justifica In addition, our other beliefs defense as possit all--but also thi certainty may in( finite knowers, t be maximally cer be more certain. What sort Self-evident am tionalists as a: reaSon that tr without ref eren us as is PCssil be true giVen . am being“ippea Ceivably be f COUI‘SeI One ma evident or inc they are, the} 88 justified by beliefs which themselves need justification, we have an infinite regress of justification. But since we cannot carry out an infinite chain of justification, that, it is argued, amounts to no justification at all. So one part of a criterion for proper basicality must preserve their foundational adequacy, yig., that they not be in need of further justification. Let us call this "epistemic finality". In addition, if properly basic beliefs are to found all our other beliefs, not only must they be as little in need of defense as possible—-preferably, they will need no defense at all-~but also they must be as certain as possible. Absolute certainty may indeed be an impossibility for humans or for any finite knowers, but certainly properly basic beliefs ought to be maximally certain, so certain that no other beliefs could be more certain. What sort of beliefs qualify, then, as basic beliefs? Self—evident and incorrigible beliefs were posited by founda— tionalists as among those which would qualify for the simple reason that they have seemed incontrovertibly acceptable without reference to any other beliefs and are as certain for us as is possible. They are the sort of beliefs which must be true given their existence. That I am in pain or that I am being-appeared—to-redly, it is claimed, could not con- ceivably’ be false for one having those experiences. Of course, one may argue that such beliefs are not really self- evident or incorrigible, but that is a separate question. If they are, then they surely qualify as properly basic. And even if one cannot them, that too is ought nevertheles tionalism seeks a: it must limit pro reasonable to acce as high a degree l and incorrigible cannot be gainsaj Plantinga of to these two c arbitrarily limi‘ accepted as bas: reasonably be ar should qualify z self-evident hay 900d argument c, of those belief for proper basi. 89 even if one cannot make a case for basing other beliefs upon them, that too is a separate question, and they themselves ought nevertheless be deemed belief-worthy. Since founda— tionalism seeks as sure and secure a foundation as possible, it must limit proper basicality to those beliefs which it is reasonable to accept without further evidence and which supply as high a degree of certainty as possible. That self-evident and incorrigible beliefs satisfy such a requirement if any do cannot be gainsaid. Plantinga objects to the limitation of proper basicality to these two categories, and undoubtedly one ought not arbitrarily limit the types of beliefs which are legitimately accepted as basic. But surely, only beliefs which it can reasonably be argued do not need evidence to be belief-worthy should qualify as properly basic. The incorrigible and the self—evident have been accepted as properly basic because a good argument can be made to the effect that the very having of those beliefs insures their truth. So if this criterion for proper basicality is sufficient, they qualify. Our thesis, then, is the following: A belief is properly basic iff (a) it stands in no need of further justification to be a rational belief and (b) it supplies maximal certainty.19 19In other words, such beliefs do not have to be absolute~ ly certain or completely indefeasible. It may be we have and can have no such beliefs. But the hallmark of properly basic beliefs ought to be that we can have no other beliefs which are more certain. why accept t properly has” me that which needs certain can be. 5 one is justified tal beliefs. Ind provides further inconsistency. '1 the more likely, wrong. Clearly, ones, and if the proportion to it less reliance tl be. But what sc beliefs can reas justification a them from such e um“ propert 90 Why accept this criterion? Simply put, because to be properly basic means to need no further substantiation, and that which needs no further substantiation is as certain as certain can be. So properly basic beliefs will be those which one is justified in having without recourse to more fundamen— tal beliefs. Indeed, every step up the evidentialist ladder provides further opportunity for error, misjudgment, and inconsistency. The more a belief is dependent upon evidence, the more likely, other things being equal, that it might be wrong. Clearly, less sure beliefs cannot ground more sure ones, and if the belief-worthiness of a belief is in inverse proportion to its height on an evidentialist path, then the less reliance there is upon evidence, the more sure it must be. But what sort of beliefs might so qualify? What sort of beliefs can reasonably be said to stand in no need of further justification and just what is it about them that delivers them from such a seemingly universal obligation? Is it a par- ticular property they possess or a particular relation they have to knowers which so distinguishes them? Surely those whose belief—worthiness is implicitly contained in their very existence qualify, but what creates that worthiness? One temptation is to say that they are their own evi— dence, but that is likely to cause more confusion than clarity. Plantinga has pointed out that "self—evident" does not mean (we might say, does not necessarily mean) that a proposition is its own evidence, but simply that it needs no evidence for warra tion that is evide (RBG, 53) But thz one step. What 5. tion belief-worth own evidence, wou tion? If not, ju Plantinga, o is one which has circumstances. I: the last sectiOn determining just or belief types Prints for humal deciding which b su‘igestion Will least 50m gene] minations? M01 Purpose of our it for true bel After all, it 11 91 evidence for warranted belief. ". . . It is . . . a proposi- tion that is evident, or known, in itself, without evidence." (RBG, 53) But that, of course, just moves the question back one step. What sort of property is it that makes a proposi— tion belief—worthy without any evidence? If it is not its own evidence, would it then be a synthetic a_prigri proposi- tion? If not, just what sort of proposition would it be? Plantinga, of course, would reply that such a proposition is one which has been designed to be so accepted in certain circumstances. In addition to the difficulties discussed in the last section, however, there is the additional problem of determining just what those conditions are and which beliefs or belief types qualify. Unless we have the epistemic blue- prints for humans or some fairly secure way of learning or deciding which beliefs qualify and under what conditions, this suggestion will be of little help. And do we not need at least some general criterion by which we can make such deter— minations? Moreover, we would have to know just what the purpose of our epistemic equipment is in order to rely upon it for true beliefs even in the most appropriate conditions. After all, it may have been designed primarily for survival, in which case truth might not always be the most advantageous or desireable acquisition. In certain cases, then, we would have properly basic beliefs which did not give us truth. Even if we could adjust for such disparities between design and current intentions, properly basic beliefs would cease to have any basic credibility, for they could as well be prohibiting us from obtaining needs as not, in W on belief at all true. But then, i if anything to 01 adequately ground rational belief, assured. By failing proper basicality alism which canr li_Z-, ground evil basic beliefs ne advantage does i potentially inff I can see. m at least the em to ground the Mthei 92 us from obtaining true beliefs in deference to more pressing needs as not, in which case they would have no implicit claim on belief at all insofar as it is the aim of beliefs to be true. But then, of course, foundationalism will have little if anything to offer as an epistemology which accounts for adequately grounded belief, whose standards must be met for rational belief, or by whose standards rational belief can be assured. By failing to measure up to our minimal criterion for proper basicality, Plantinga’s proposal creates a foundation— alism which cannot do what foundationalism is meant to do, yiz., ground evidential beliefs. It cannot do so because its basic beliefs need grounding themselves. In that case, what advantage does foundationalism give us over coherentism or a potentially infinite chain of justification? None as far as I can see. Vis—a-vis the former, it may even be weaker, for at least the entire coherent system can be thought in a sense to ground the individual beliefs of which it is composed. Vis—a—vis the latter it seems to fail to give us an adequate reason to stop at any particular place. If at least some basic beliefs are defeasible, why not all? If there is a basis for a distinction between defeasible and indefeasible basic beliefs, why not simply call the latter basic? Would not, in other words, such a basis he a de facto criterion for proper basicality? If all basic beliefs are defeasible, how do we prevent an infinite regress, or what advantage does foundationalism have over that position? both. Regardles if basic beliefs dationalist noe grounded will h. epistemic finall beliefs must ha! belief or belief and finality mus but Plantinga's for such a prop 93 So perhaps we cannot say exactly what it is that creates the intrinsic belief-worthiness of properly basic beliefs. For some it may be a property of the belief itself, for others a property of the believing experience, and for still others both. Regardless, it cannot be an external set of conditions if basic beliefs are to ground all other beliefs in a foun- dationalist noetic structure, for beliefs which are so grounded will have neither the maximal certainty nor the epistemic finality which foundationalist properly basic beliefs must have. Whatever property or properties of a belief or belief producing experience entail such certainty and finality must be intrinsic to that belief or experience, but Plantinga’s largely externalist position does not provide for such a property. - .- r - , a c. .. mun“ , _' . .5 . , a,“ minim-no 3.0a Isa-19m mu ad mm :1 ,lII-IM“ " h “I ._ ' I .guoi a at aleiled 393:0 ifs bnxtowp a: all atotlod stand-fie E: MW 4.-.; _ -_ ; ,-.-._. --'. ,1,— '. -3:- : '::: -I. ' '- ._ .-. .Em E53013” I Plantinga . .. . .. .~ 1 . '... .-,- .-.---§..m-.-.-s_:_p propriate circumi L * of this claim I circumstances ex? to marshall evid rational. The pi what beliefs arl how according t1 rational noetic properly basic properly basic ‘ foundationalism argue that eve] basic beliefs, Significantly : Section ] Plantinga pro} i belief does no i 1Will endeavc Plantinga pro} an eVidence es CHAPTER IV THEISM, PROPER BASICALITY, AND JUSTIFICATION Plantinga has argued that belief in God can, in ap- propriate circumstances, be properly basic. The importance of this claim lies in the fact that if true, when those circumstances exist, the Theist is not under any constraints to marshall evidence in behalf of Theism for his belief to be rational. The preceeding chapter sought to make clear exactly what beliefs are or can be properly basic for Plantinga and how according to him such beliefs receive their warrant in a rational noetic structure. I then argued that Plantinga properly basic beliefs do not measure up to the requirements properly basic beliefs need to meet if the epistemic goals of foundationalism are to be attained. In this chapter, I shall argue that even if we accept Plantinga’s notion of properly basic beliefs, the rational fortunes of Theism are not thereby significantly improved. Section IV.1 will argue that construing 'Theism1 as a Plantinga properly basic belief instead of an evidential belief does not markedly improve its rational acceptability. I will endeavor to show that construed as any Viable type of Plantinga properly basic belief, Theism easily devolves into an evidence essential belief, so that the claim of its proper 94 l i Hammett. ans .nmxua Ham .mm _ A - W . l, ‘- seamen-.3 «11 -5. ,-.- _,1,-;-; -‘.,---.'_'- '- - t a. ' «=...' mgr-tuna!!! i basicality, as vi ._ -.... . r-'-.?.-'~i'x'-.',v_-’JC; i in warranted beli - .~ ' belief in God pg; even were it cl section IV.4 it v which, in chaptel position, resul the rationality provide. Finall severe difficul whether or not (b) if so, undel that if those 6 cality of Thai: 1v.1 mums Suppose I 1 authority beli lBY “aut authority . detail 95 basicality becomes moot. Section IV.2 will argue that Plantinga’s view of proper basicality does not help us determine whether or not belief in God is properly basic, and that even if he is correct, we cannot determine the rational- ity of belief in God by means of its basicality alone. Section IV.3 will contend that Plantinga’s View of proper basicality, as with any "reliabilist" position, results not in warranted beliefs, but in warranted believings, so that belief in God per se would not ipso facto be a rational belief even were it clearly to be a pmoperly basic belief. In section IV.4 it will be claimed that the internalist elements which, in chapter VI, I will argue are implicit in Plantinga’s position, result in "type—token" ambiguities which undermine the rationality which the proper basicality of Theism was to provide. Finally, in section IV.5 I will argue that there are severe difficulties standing in the way of determining (a) whether or not belief in God is Plantinga properly basic and (b) if so, under what conditions. Moreover, it will be argued that if those difficulties can be overcome, the proper basi- cality of Theism will be a moot point. IV.1 PLANTINGA PROPERLY BASIC BELIEFS AND RATIONAL THEISM Suppose Plantinga is right. Suppose memory beliefs, authority beliefs,1 and many observation beliefs are properly 1By "authority belief" is meant a belief taken on authority. Chapter V will discuss authority beliefs in detail. s .m u at! _m13 131.8591!“ =831 ”to an” Yd baa of sum i '3»; .,; ... IBQO‘I-l 1:: WHZ‘J 2‘1.9fl."3'_'.|"-. ': T-i'i)‘ "rt-Jae? ' (in: C .VI '91“ .. - wmidWyeheli -,-.. . :. ,’:"'J' "-'7-' :; "2‘ -'. '=. ' .9... -.' . -:r-' ritual-.6 whattype of P, -.---.-.' ni‘ God In, an anti belief, or some ity, observatii~ types of Plant is some other w M basic we will be an But there is o generis belie beliefs. The: Theism as a le evaluate the belief, an at these seem ti it as either : require the u to substanti as basic for belief that ' Promising, 96 basic for me. Moreover, suppose the rationale he proposes for their proper basicality is correct. That is, suppose there is some set of circumstances which warrant my having such beliefs without evidence, providing, of course, my epistemic apparatus is working properly. Under what sort of conditions would my belief in God be properly basic? In other words, what type of Plantinga properly basic belief would belief in God be, an authority belief, a memory belief, an observation belief, or some other sort of Plantinga basic belief? Author— ity, observation, and memory beliefs are the most conspicuous types of Plantinga properly basic beliefs, but perhaps there is some other way Theism could be basic, for instance as a ggi generis basic belief. Of course, if it is absolutely unique, we will be unable to say much of anything at all about it. But there is one class of beliefs under which a relatively ggi generis belief might well be classified, namely, innate beliefs. Therefore, in examining the implications of taking Theism as a legitimate Plantinga properly basic belief, I will evaluate the epistemic results of taking it as an innate belief, an authority belief, and an observation belief, for these seem the most hopeful. It will be argued that taking it as either an innate or an authority belief would eventually require the use of observation beliefs by the believer himself to substantiate that status, thereby undermining its status as basic for him. Thus, it is as a properly basic observation belief that Theism’s prospects for proper basicality are most promising, but, I shall contend, here too the promise V " ’ = e- -1“. . .l - '. LIV“; Fl ” - .' ‘w‘. 5“ A . ' a.) I ..”...g, , s? .. “I ...-.5- “fibfi-{frfi' ‘3‘ 1 ” e'mfitvofq \m , Inciakbnee to :10: Judy mohnu 5119qugq sniimr' u£f_ “f i . . argued p '7 3.5th :- . C‘- | at; I. r‘ -- . 4.1.. i .amow radio :13 . -; - -. i .1.’ 14‘ and ‘1!" I r nil-vet _=v:: oqy: 3|. ‘ suchasmightbei ; ~ ._ 22."! so: ing philosopherg l'."-. . .1 exists. Perhap‘ God at a certail of ticks on sow starts ticking " i of brain develt i Reformed theol I ' . structed that exists. Expou Were it I human bei and with believe external human co: does not substand believe a clever feelings Thus: if fum beholds the s tricacies of created exis. 97 outstrips the reality. IV.11 Belief in God as an Innate Belief It is at least conceivable that belief in God is simply a valid innate belief, programed in by God so that unless our cognitive faculties are adversely altered by deleterious input such as might be garnered from some epistemically dysfunction- ing philosophers, one will simply have the belief that God exists. Perhaps we haVe a genetic code that causes belief in God at a certain stage of development or at a specified number of ticks on some genetically engineered internal clock which starts ticking at conception, birth, or some definite stage of brain development. Plantinga, following Calvin and other Reformed theologians, suggests that perhaps we are so con- structed that in certain conditions we simply believe God exists. Expounding Calvin’s position, he writes, Were it not for the existence of sin in the world, human beings would believe in God to the same degree and with the same natural spontaneity that we believe in the existence of other persons, an external world, or the past. This is the natural human condition; . . . The fact is, . . . one who does not believe in God is in an epistemically substandard position-—rather like a man who does not believe that his wife exists, or thinks she is like a cleverly constructed robot and has no thoughts, feelings, or consciousness. (RBG, 66) Thus, if functioning normally (i.e., as designed), when one beholds the starry skies, marvels at the complexities and in~ tricacies of nature, or thinks about the contingency of created existence, one naturally, spontaneously and properly believes in God. we 3891M is!!! ca n03 yd ni hams-word Emir-l M fimi 3”an t-zifl-I‘fzi'3filLi3 {muslin ~._- Lia-.11 -+-'-.= -.'-'.:-.' 291510332 . ' ' ifl-et-therwe 1 Ir ' E —n '_.—. 3: u 'a ' --';r;' ..-" ' - : .--. f-z'J' ments: as “ us--somehow—-or 1‘ I - he; pa! were the case, I -._.‘- my letter, an cheer: ties will be (11': There is, 1 the belief that be innate. But cation, and a 3'1 ideas were jus I such a regress ‘It is in; to ask how one such circumst: directly, jus spontaneously I could be cer in me a belief validity of s does one know that I have so validity than Ibelieve I a1 ls valid. M exception, an by merely no reaction. 01 logical dist: external wor] Such as thos: God. 98 Suppose that in fact this is the case. Would such an innate belief constitute knowledge?2 Surely, we could not intuitively know it did, for then one could reasonably be asked to justify (i.e., present evidence for) the validity of that intuition. It would seem we could only know it is true if either we learned it on authority--God might reveal it to us somehow--or we discovered it empirically. If the former were the case, it would be an authority belief, and if the latter, an observation belief (at best). These two possibili- ties will be discussed in the succeeding sections. There is, however, one other possibility. Theoretically the belief that belief in God is innate and valid could itself be innate. But then that belief would be in need of justifi- cation, and a justificatory regress would begin wherein innate ideas were justified by innate ideas. The only way to stop such a regress would be to show that (some) innate ideas are 2It is important to note that this question is not meant to ask how one would know he spontaneously believed in God in such circumstances. That may well be something he knows directly, just as I know that when appeared—to-treely I spontaneously believe I am seeing a tree. By introspection I could be certain that under some conditions there welled up in me a belief in God. Rather, this question is aimed at the validity of such a spontaneous belief. How, in other words, does one know if and when it is belief-worthy? The mere fact that I have such beliefs is no more reason to believe in their validity than the mere fact that when I am appeared—to-treely I believe I am seeing a tree is reason to believe that belief is valid. Most often, of course, it is, but not without exception, and one does not distinguish between these cases by merely noting the naturalness and universality of the reaction. Of course, it is also important to note that the logical distance between appearances and beliefs about the external world is much shorter than that between experiences such as those noted in the text and the resultant belief in God. true because or 5 would amount to 5 basic in the 018 clearly to be tht of the prone—Nie anywhere near tr course does not Perhaps inr necessary truth ideas which inc thatany additic although Theism adequate additi larity, be used an argument mic Obviously, sin GOd exists, th The real substantiate i InnateneSS I Si not entail tn aDPeal to in Sub'Stantiate 99 true because or as a result of being innate. That, however, would amount to a proof that (some) innate ideas are properly basic in the classical foundationalist sense. But it seems clearly to be the case that neither innateness itself nor any of the properties innate ideas might possess entail truth to anywhere near the degree CF basic beliefs do, and so such a course does not appear promising. Perhaps innateness plus some other premises entails the necessary truth of innate ideas (or some subset of innate ideas which includes Theism). But (1) it seems implausible that any additional naturalistic premises would do so, and (2) although Theism or some inference from Theism would supply adequate additional premises, they cannot, on pain of circu— larity, be used as an argument for the truth of Theism. Such an argument might, for instance, look like this: (1) Man was designed by God. (2) God’s design includes innate ideas. (3) Innate ideas included in the design are true. (4) Belief in God is an innate idea. Therefore, belief in God is true. Obviously, since premise (1) assumes as well as entails that God exists, the argument is viciously circular. The real difficulty is that no matter how we try to substantiate the truth of belief in God by virtue of its innateness, such an argument will either beg the question or not entail the desired conclusion. If, on the one hand, we appeal to innateness itself or use Theistic premises to substantiate the validity of innate beliefs, we beg the question (at least in regard to the validity of Theism); if, on the other han to have a nonevi Suppose, f< which showed tha other innate bel experiment would for Theism migh‘ study could only verifying them 6 not be empirica haps, lend at 1‘ is indeed an e innateness of T Naturally, ideas are or a m evidence able to accept conclude that, second, the fac 100 on the other hand, we appeal to empirical evidence, we cease to have a nonevidential belief. Suppose, for example, we conducted an empirical study which showed that Theism is an innate belief and that all our other innate beliefs are true. First, even such a successful experiment would not itself entail that belief in God is true, for Theism might be the one exception. Since the empirical study could only show that all other innate ideas are true by verifying them empirically, the fact that belief in God could not be empirically verified by the same methods would, per- haps, lend at least some credence to the possibility that it is indeed ani exception. Alternative explanations for the innateness of Theism would only strengthen our suspicions.3 Naturally, empirical evidence that all other known innate ideas are or at least tend to be true would give us prima facie evidence that belief in God is true, making it reason- able to accept it. Indeed, it would then be reasonable to conclude that, prima facie, it is properly basic. But, second, the fact that it was established as properly basic by virtue of evidence would make it an evidential belief for us. In other words, we face a dilemma. If Theism is properly basic by virtue of being a designed, innate idea, given reason 3Of course, no such evidence for the truth of all our innate ideas exists; indeed, no compelling evidence for even the existence of substantive innate ideas exists. Moreover, it should also be noted that if, contrary to the above scenario, it could be plausibly demonstrated that belief in God is innate, but as it happens the only innate idea, the above empirical argument would not even be possible. to doubt the tru Particular, its ‘ for the truth of ular. But then In regard ‘ propositions, "I likely true)" 5 maximally certa. it is Plantinga which make it i] innate belief Whenever signi‘ belief only if innate and the course, it bec. For clari 1. Assu Virt 2. Asst true a. b. C. 3- Bel I h is 4‘ Suc me are 101 to doubt the truth of innate ideas in general or Theism in particular, its credibility can only be restored by evidence for the truth of innate ideas in general or Theism in partic- ular. But then it becomes an evidential belief for us. In regard to Theism, the reason for this is that the propositions, "Innate ideas are properly basic (and therefore, likely true)" and "Belief in God is innate," are neither maximally certain nor epistemically final. In other words, it is Plantinga's construal of basic beliefs as defeasible which make it impossible for Theism taken as a properly basic innate belief to maintain its epistemic status as basic. Whenever significantly challenged, it remains a justified belief only if it can be demonstrated both that Theism is innate and that as such it is (likely) true. But then, of course, it becomes an evidential belief. For clarity, let us schematize the argument. 1. Assume belief in God is properly basic by Virtue of being an innate idea. 2. Assume that one or more of the following is true: a. I have reason to doubt God exists. b. I have reason to doubt "God exists" is innate. c. I have reason to doubt innate ideas are true by virtue of their being innate. 3. Belief in God is no longer basic for me unless I have evidence that defeats each of a-c that is true of me. 4. Such evidence will justify belief in God for me in accord with one or both of the following arguments: a. 1) Evidence E shows that God ex— ists. Therefore, God exists. b. 1) Evidence E shows innate ideas 5. Belief me by Note that even i: on direct evide] do so only beca But surely evid innate ideas a1 God’s existence in God is an ev It appears basic innate b doxastically h Will clearly ‘ Support. So 1 basic in a way must be so as 102 are (highly likely to be) true. 2) Evidence E’ shows belief in God is innate. 3) Therefore, belief in God is (likely) true. Therefore, God exists. 5. Belief in God is now an evidential belief for me by virtue of argument a, b, or both. Note that even if I continue to believe in God without relying on direct evidence for God’s existence itself, I rightfully do so only because I have evidence that this is justified. But surely evidence that belief in God is innate and that innate ideas are (likely) true is (indirect) evidence for God’s existence. Clearly, then, in such circumstances belief in God is an evidential belief.‘ It appears, then, that taking belief in God as a properly basic innate belief will not establish its basicality in a doxastically helpful way, i.e., such that in most cases it will clearly' be a :rational belief without any' evidential support. So it would seem that if belief in God is properly basic in a way that will be of doxastic benefit to Theism, it must be so as either an authority or an observation belief. IV.12 Belief in God as an Authority Belief Suppose, then, that belief in God is properly basic because it is an authority belief. It would seem that then 4Moreover, it is obviously the case that there are substantial reasons to doubt the innateness of belief in God, that innate ideas are true, and that God exists, and so assuredly there is need of some evidence for either the first two or the third of these propositions. its proper basic: limited noetic Si and all rational appear those mc basically on aui question authori authority in Vi belief would ce‘ i.e., it would c now be incumben authority in vi But how 02 compilation of unless the ant] sense of prope alone. Moreow an authority, that shows be pain of an i: 103 its proper basicality would only extend to people in rather limited noetic situations. Children, people isolated from any and all rational discourse about the gods, and such like would appear those most clearly justified in believing in God basically on authority. But as soon as they had reason to question authority beliefs in general or the validity of that authority in virtue of which they believed in God, their belief would cease to be justified by that authority alone; i.e., it would cease to be a basic authority belief. It would now be incumbent upon the believer to at least justify that authority in virtue of which he believed. But how can one verify or confirm that someone or some compilation of writings is an authority on God? Clearly, unless the authority of some authority is foundational in my sense of proper basicality, it cannot be done by authority alone. Moreover, as long as evidence is necessary to validate an authority, one will eventually have to come upon evidence that shows belief in that authority to be justified. Upon pain of an infinite regress or circularity, that evidence cannot simply be other authorities which lack intrinsic belief—worthiness. So, if some authority is not intrinsically belief-worthy, at least some of that evidence will need to be observational.5 In other words, there will have to be obser— vational beliefs which provide evidence that the foundational 5Of course, it is arguable that God’s pronouncements are intrinsically belief-worthy (i.e., that "What God says is true" is properly basic), but we would still need some confirmation that we were being addressed by God. authority is prOE in its pronounC‘ examine the implf basic observatiO 1v.13 relieLD We have an is properly basi sense, it is as large and deep which claim to Objections abc nonphysical (a1 , the privacy of and such like. to deal with a to this proble beliefs in gel the latter 0: observation b Planting beliefs as, n I haVe done," (RGB, 31) ml eaCh Such be: conditions C Such a rule 104 authority is properly foundational or that it is indeed right in its pronouncements. Given these difficulties, let us examine the implications of taking belief in God as a properly basic observation belief. IV.13 Belief in God as an Observation Belief We have arrived at the conclusion that if belief in God is properly basic in any doxastically or epistemically helpful sense, it is as an observation belief. It is, of course, a large and deep question whether or not any or all experiences which claim to be experiences of God actually are veridical. Objections abound, centering upon such problems as God’s nonphysical (and therefore nonphysically observable) nature, the privacy of such experiences, God’s purported ineffability, and such like. It is beyond the scope of this dissertation to deal with all the arguments and counter—arguments related to this problem, but we can examine the nature of observation beliefs in general and belief in God in particular to see if the latter can plausibly be accepted as a properly basic observation belief. Plantinga, we noted above (pp. 60—61), holds that such beliefs as, "God created all this," "God disapproves of what I have done," "God forgives me," and "God is speaking to me," (RGB, 81) may be properly basic. Suppose they are. Then for each such belief there is some rule R which states that under conditions C that belief is properly basic. For instance, such a rule might go as follows: It should I being apprised informed that h be the case, r source of propt basic belief w) need not be con God disapprove well trigger b previously bel \ 60f cours. above, but th PIOperly basin eXistent belie sP‘ecifxinq t1 disapproves o: 105 R1 In the following conditions, humans will have the properly basic belief that God disapproves of what they have done: Cg Their cognitive mechanisms are working properly; C5 Their moral awareness, feelings, and sensibilities are functioning up to speci— fications; Cg They have committed some moral offense against God's will; C,: They are apprised.that.they'have committed such an offense. It should be noted that it need not be supposed that in being apprised of having committed this offense, one need be informed that he committed it against God. Perhaps this will be the case, but if such a situation is to function as a source of properly basic belief in God (or of that properly basic belief which directly implies belief in God), then one need not be convinced of God’s existence before believing that God disapproves of what he has done. Such a situation could well trigger belief in God’s disapproval even if one had not previously believed in or even thought about God.6 6Of course, one could add belief in God to the conditions above, but then one could not argue that belief in God is properly basic by virtue of this rule because explicit and existent belief in God would be presupposed in the conditions specifying the proper basicality of believing that God disapproves of one’s action. The conditions specified in the above rule supply the warrant for my properly basic belief, but if those conditions include believing in God, then I must already believe in God for those conditions to apply. In that case, I can hardly use the belief warranted by those condi— tions as a premise for the immediate implication that God exists, nor need I. In fact, if such were the case, it would be more plaus- ible to argue that belief in God’s disapproval of what I have done is a direct implication of my prior belief in God. Here one might add that such an implication follows from a partic— ular conception of God, and that belief that God disapproves of what I have done requires a previous conception of God which would entail His disapproval of certain acts. Of ' ...... '1 There are, 3 exist if belief or authority bel 1 "God is disappr belief D), is pr for it to be pro it properly bas proper conditic al belief. But now su evidence. Foul of it. First, D is properly however, that beliefs are Pl ible. If D is will have to \ Course! While beliEf in Go, Plantinga has A_Supports B either. Ther that God dise beliefs are m were one of 1 t° be Warrant by this rule. belief in Sec p?“ a Prior simultaneousj §ltuation_ Implicit in, was accepted belief that 106 There are, however, complications similar to those which exist if belief in God is considered a properly basic innate or authority belief. For Plantinga, that such a belief as, "God is disapproving of what I am doing," (let us call it belief D), is properly basic need not be known by the believer for it to be properly basic for him. It is sufficient to make it properly basic for him that he have the belief under the proper conditions. Being properly basic, it will be a ration- al belief. But now suppose that belief is challenged by conflicting evidence. Four sorts of arguments can be mustered in behalf of it. First, one can demonstrate that in those conditions D is properly basic. Given Plantinga's view of basicality, however, that will not in itself count for much, for many beliefs are Plantinga properly basic which are easily defeas— ible. If D is one of the less defeasible, that, of course, will have to be demonstrated also. But then it would seem course, while all these beliefs would be mutually supportive, belief in God would not thereby acquire any warrant, for Plantinga has rightly argued that while A warrants B only if A supports B, that A supports B and B A does not warrant either. Therefore, belief in God is not warranted by belief that God disapproves of what I have done just because these beliefs are mutually supportive. Naturally, if belief in God were one of the conditions in the above rule, it would have to be warranted by some other situation than that warranted by this rule. But as stated above, it is not necessary that belief in God's disapproval of what I have done need presup— pose a prior belief in God. Both beliefs might be triggered simultaneously upon finding oneself in the appropriate situation. Indeed, belief in God itself might be completely implicit in such an experience so that conscious belief in God was accepted by (conscious or unconscious) entailment from belief that God disapproves of what I have done. W1 ll-C . .". .__l’ . . _ $ ' m ‘ N my belief of D I) because I bel properly basic addition, the e in D. believing basi Second , ed. For insta accounted for stances could have no more 3' demonstrating effect as wel : Third, sufficient w Thus, in re 107 that D’s basicality is supported by evidence, evidence that it is basic and evidence that its basicality is strong enough to withstand whatever evidence is marshalled against it. Now my belief of D will no longer be properly basic, for I believe D because I believe it is properly basic and I believe it is properly basic because I have an argument to that effect.7 In addition, the evidence that D is basic will play an essential role in my belief that D is basic and, therefore, in my belief in D. While D might remain a valid belief for me, it is clearly no longer for me a belief which I am warranted in believing basically. It is now an evidential belief for me. Second, the challenging evidence itself could be defeat- ed. For instance, an alternative theory which nontheistically accounted for my coming to belief in God in those circum— stances could itself be shown to be untenable, or at least to have no more justification than belief in God as basic.8 But demonstrating this would, of course, require evidence to that effect as well as argumentation. Third, the basic belief itself could have, as basic, sufficient warrant to itself defeat the proposed defeater. Thus, in response to this sort of criticism from Quinn, Plantinga writes, Two questions . . . arise. First how does the 7Of course, that argument will need to present evidence that D is indeed basic, for it is difficult to imagine how the proper basicality of such a belief could be demonstrated without some appeal to evidence of some sort. aSee Plantinga, "The Foundations of Theism," Faith and Philosophy, Vol 3, No. 3. July 1986, pp. 306—312. '0 I .A . RV‘ . ,,.q.y ;‘ stitutes . 49-1 n. .- secondly, 'g ' these defe 1' . - 4.5 a of 10” t existence defeater f er—defeat: these pro- that the defeater to have : reliable argument experts are evenl I have 0 defeater theistic ity .’ last claim, n2 tinue to acce' it seems this have made it which suppor simple direc certainly ev its complete 5 direct evide results sho as evidence less of hi 108 degree of nonpropositional warrant enjoyed by your belief in God compare with the warrant possessed by the alleged potential defeater? It could be that your belief, even though accepted as basic, has more warrant than the proposed defeater and thus con— stitutes an intrinsic defeater-defeater. . .. And secondly, are there any extrinsic defeaters for these defeaters? Someone argues that the existence of 1013 turps of evil is inconsistent. with the existence of (kfih I may then have an extrinsic defeater for this potential defeater. This defeat— er—defeater need not take the form of a proof that these propositions are indeed consistent; if I see that the argument is unsound, then I also have a defeater for it. But I needn’t do even that much to have a defeater. Perhaps . . . I learn from reliable sources that someone else has shown the argument unsound; or perhaps I learn that the experts think it is unsound, or that the experts are evenly divided as to its soundness. Then too I have or may have a defeater for the potential defeater in question, and can continue to accept theistic belief in the basic way without irrational- ity.9 Now all that Plantinga says here seems right except his last claim, namely, that given these responses one "can con- tinue to accept theistic belief in the basic way." Indeed, it seems this is just what one does not do, for the challenges have made it necessary to add beliefs to my noetic structure which support my belief in God. They are not, to be sure, simple direct evidence for the truth of Theism, but they are certainly evidence that Theism is not false, and so part of its complete justification for me. Indeed, they count as in- direct evidence for Theism in the same way that experimental results showing the inadequacy of Newtonian Mechanics count as evidence for the truth of Einsteinian Mechanics. Regard- less of which of Plantinga’s alternatives is adequate, given 91bid., pp. 311—312. , defi'eater‘srseems ' longer .sismply . by virtue of h the appropriate because one ha- given this par for Theism, th rise to Thais. doing") is its ing evidence 1 evidential bel In each . me the moment proper basical a potential de Such wou God itself we ist sense. T outside the basicality of construing p the directio belief. Trad 109 knowledge of the defeaters, knowledge of the defeater- defeaters seems necessary. If so, then belief in God is no longer simply basic, believe—worthy in its own right merely by virtue of having come about in the appropriate fashion in the appropriate circumstances. It is now rationally believed because one has evidence against potential defeaters which, given this particular epistemic situation, counts as evidence for Theism, thereby making the latter an evidential belief. A fourth response would be to argue that a belief giving rise to Theism such as D ("God disapproves of what I am doing") is itself true. That, of course, would mean present- ing evidence in behalf of D, and clearly D would then be an evidential belief simpliciter. In each case, then, D becomes an evidential belief for me the moment I defend it, whether that defense is of its proper basicality, its truth per se, or its viability despite a potential defeater. Such would not be the case, however, if D or belief in God itself were properly basic in the classical foundational- ist sense. Then appeal would not have to be made to anything outside the belief itself, such as evidence for the proper basicality of the belief or for the belief itself. Indeed, construing properly basic beliefs as Plantinga does inverts the direction of argument for the rationality of a basic belief. Traditional, classical foundationalist properly'basic beliefs are considered properly basic because they are, allegedly, irrefragably true. Their proper basicality and I ~ and mtionahity from the belie warranting ord ference between classical founc properly basic truth, but to their truth. If it is God possesses would seem ex sense that my on a par with of fact, no beliefs. Sur belief facin Theism faces basicality u that Theism Plantinga pr 110 belief-worthiness follow from their undeniable truth and consequent rationality. Contrariwise, the belief-worthiness and rationality of Plantinga properly basic beliefs follow from the belief's demonstrated proper basicality. Such a warranting order seems not only to mark a significant dif- ference between Plantinga properly basic beliefs and those of classical foundationalism, but also to insure that Plantinga properly basic beliefs will become evidential beliefs once they come up against a significant challenge. For then one must appeal not to their incorrigibility or self—evident truth, but to the evidence for their proper basicality or their truth. If it is this sort of proper basicality which belief in God possesses, its rationality by virtue of that basicality would seem extremely weak. It would be basic only in the sense that my other observation beliefs are properly basic, on a par with memory and authority beliefs, and, as a matter of fact, more defeasible than many if not most of those beliefs. Surely, to remain rational any observation or memory belief facing the sort of difficulties and problems which Theism faces would require more than an appeal to its proper basicality under appropriate conditions. Since even the claim that Theism is properly basic faces difficulties most other Plantinga properly basic beliefs do not, the belief-worthiness derivable from this claim becomes even more tenuous. possible so the - the sorts of -: While some of sipate, closer most obvious p an observation physically obs at least nonph Two possibili . generis or it makes it poss‘ the convictio thinks about :‘ Here, sin appear. If e: be some evid Moreover, th on authority outlined in 111 IV-14 WW Suppose, however, that direct observation of God is possible so that one did not at all infer God’s existence from the sorts of beliefs considered basic in the last section. While some of the problems mentioned above might seem to dis- sipate, closer examination reveals this is not the case. The most obvious problem is specifying just what would count as an observation of God. Since by definition God is not physically observable, there must be some sort of "inner" or at least nonphysical means of perceiving or experiencing God. Two possibilities exist: Either such an ability is sui generis or it is a function of whatever faculty (or faculties) makes it possible for us to have analogous experiences, like the conviction one has of the truth of modus ponens when one thinks about it. Here, similar difficulties to those mentioned above again appear. If experience of God is sui eneris, then there must be some evidence that it is for us to rationally so believe. Moreover, this evidence must either be observational or based on authority. Either way, we face problems similar to those outlined in the last 2 sections. If, on the other hand, it is classed with rational intuitions, one would have to account for the fact that while most anyone can be brought to the place where they have rational intuitions about such things as modus ponens, very few have Theistic confirming experiences which produce the same level of conviction. This and other asymmetries may not argue conclusively against the belief-worthines for one who has i be summarily di“ inclusion of Th‘ at least boarde appeal to evide evidential beli IV.15 Conclusi If there which immediat belief in God i tion of the no which has bee] gain for Their basic, i.e., 1 God is just a a properly b. Theist who be Perience of n his belief i 112 belief-worthiness or the proper basicality of belief in God for one who has had such an experience, but surely they cannot be summarily dismissed. Moreover, once one argues for the inclusion of Theistic intuition with rational intuition, one at least boarders on the use of and perhaps inevitably must appeal to evidence. That is, Theism once again becomes an evidential belief. IV.15 Conclusions If there are direct experiences of God or experiences which immediately entail the existence of God, then surely belief in God is Plantinga properly basic. But if the evalua- tion of the noetic status of Plantinga properly basic beliefs which has been made above is correct, this is not much of a gain for Theism. Since Plantinga's basic beliefs are weakly basic, i.e., highly defeasible, it would seem that belief in God is just as vulnerable as it would be without calling it a properly basic belief in these circumstances. Does the Theist who believes in God because he has had a supposed ex— perience of God lose anything in relation to rationality if his belief is not considered properly basic, but instead as one whose rational justification is evidentially grounded in that experience? We will argue below (chapter VI) that insofar as Plantinga properly basic beliefs are grounded in experience, that experience can be construed as evidence for them. After all, when any Plantinga properly basic observa- tion belief is challenged, the circumstances that ground or warrant that bel \ that belief. 5' a belief as Pro] the noetic situ the belief? If it becomes an e If such 1‘ status as pr0p tradition of t i quently and fo properly basic ing confidence per basicalitj it a viable, a reasons consi argument), re experience, i that belief Simply to st Compelling 01 itself, one for this cla belief . 1° 113 warrant that belief as properly basic then become evidence for that belief. So construed, the only difference between such a belief as properly basic and such a belief as evidential is the noetic situation of the believer. Has he reason to doubt the belief? If not, it remains properly basic for him; if so, it becomes an evidential belief for him. If such is the case, belief in God has surely lost its status as properly basic for most of us in the intellectual tradition of the West, for that belief is challenged so fre- quently and forcefully that simply responding that it can be properly basic for some people goes little towards resurrect- ing confidence. Surely, most Theists don’t appeal to the pro- per basicality of Theism, but to some reasons why they feel it a viable, attractive, or compelling belief. Whether these reasons consist of logical proofs (such as the ontological argument), revelations (as, e.g., the Bible), or religious experience, in the context of justification, they are evidence that belief in God is warranted. Even if one’s defense is simply to show that none of the antitheistic arguments is compelling or that none are any more belief-worthy than Theism itself, one will necessarily mount a great deal of evidence for this claim, again making Theism an evidentially supported belief.10 10Evidence may not be required for counters to some (perhaps all) of the strictly philosophical objections to Theism, but it surely‘ would be for replies to arguments against Theism arising from biblical criticism, history, psychology and psychoanalysis, biology, anthropology, physics, neuroscience, etc. Even if most or all of the antitheistic arguments from these areas include philosophical principles, The only W ficantly PrOPeI it were thereb? for which I h require an ovei God in jeopan truth. But whi of God might , belief, it is sort of episi properly basic the evidence require that antitheistic 110f cour states or s fortitude tha basic beliefs are none, how if there exis evidential b If there are foundational than any oth« evidence, bu But th Evidence whj Either be (3. Will haVe an 0f luStifyir than any oti’ Theism in a no less war 9hapter VI Jl‘lSt as Nat Wlll eVentu can Present many Signif wlll haVe a 114 The only way belief in God would be helpfully and signi- ficantly properly basic in our noetic situation would be if it were thereby to meet the criterion for proper basicality for which J: have argued in Chapter III.“ It would then require an overwhelming amount of evidence to place belief in God in jeopardy, for the belief itself would "entail" its truth. But while belief in God by virtue of direct experience of God might conceivably count as a Plantinga properly basic belief, it is difficult to imagine how it could ever have the sort. of epistemic status that classical foundationalism’s properly basic beliefs have (and a fortiori, for all the other the evidence which is used antitheistically will usually require that it and/or other evidence be used in the anti- antitheistic arguments too. 11Of course, perhaps no beliefs, not even self-presenting states or self-evident truths, have the epistemological fortitude that I have argued must be the case with properly basic beliefs if foundationalism is to be tenable. If there are none, however, then the claim that Theism is rational only if there exist cogent arguments from foundational premises to evidential beliefs which support Theism falls to the ground. If there are no properly basic beliefs, then there are no foundationally grounded beliefs and Theism is no worse off than any other belief in that regard. One might still require evidence, but it will not be foundationally grounded evidence. But then what type of evidence will be required? Evidence which is not grounded in foundational beliefs will either be grounded in a coherentist noetic structure or it will have an infinite justification path. If infinite strings of justifying beliefs are permitted, Theism seems no worse off than any other belief. But likewise, if one need only ground Theism in a coherentist noetic structure, then too it seems no less warranted than other beliefs. If the arguments in chapter VI are correct, Theism will be warranted by Theism just as Naturalism will be warranted by Naturalism, for both will eventually fall back upon themselves. As long as Theism can present a coherent and equally complete (i.e., handle as many significant beliefs as its rivals) View of the world, it will have as much warrant as its alternatives. ways in which i IV.2 W As we haw mitted from 01 situation to a paratus is fun designed. Af sense of the < I shall argue a nontheistic functioning it worthiness, c I shall then with design < such functi01 115 ways in which it might be Plantinga properly basic). IV.2 DESIGN AND WARRANT As we have seen, Plantinga holds that warrant is trans- mitted from one belief to another or from an experiential situation to a belief only when a cognizer’s epistemic ap- paratus is functioning properly in conditions for which it was designed. After a brief attempt to show how one can make sense of the concept of "design“ in a nontheistic framework, I shall argue that the sorts of difficulties one might, from a nontheistic framework, raise against the view that "proper functioning in accord with design" necessarily entails belief- worthiness, can also be raised within a Theistic framework. I shall then contend that, given proper functioning in accord with design does not entail the truth of beliefs produced by such functioning, it is necessary to know the following: (1) What the epistemic "design plan" is. (2) In what circumstances and to what extent that design plan is aimed at producing true beliefs. (3) For any particular belief, how the epistemic design plan applies to it. But, I shall claim, if we can determine the answers to these questions, the basicality of belief in God will be of no epistemic significance for the question of Theism’s rational- ity, for we will be able to answer that question whether or not belief in God is basic. That is, whether or not belief in God is basic, we can only know it is rational if we can show God exists by some other means. The concept of design makes straightforward sense within a Theistic fram sarily presuPPc of a naturali: designing SYS'Ce engineering r113 These rules ha‘ varying enviro others not. having been "i laws to functi during the 6 course, this produced an < particular en minimal sense a proper way Of course, c will likely, 1y that our true beliefs ti‘mary proc "designed" t a certain le bring about moment I prc identified design in a 116 a Theistic framework, but Plantinga feels it does not neces- sarily presuppose one, and this seems right. One can think of a naturalistic evolutionary process as an impersonal designing system. Evolutionary laws of development become the engineering rules by which organic mechanisms are "designed". These rules have been applied to organisms within various and varying environments, some surviving the "prototype process," others not. Those which do survive can be thought of as having been "impersonally designed" by means of evolutionary laws to function in those environments which the laws governed during the evolutionary process. Nonmetaphorically, of course, this simply means that the evolutionary process has produced an organic mechanism which is able to survive in particular environments by means of various systems. In that minimal sense, all can agree that our cognitive faculties have a proper way of functioning in certain conditions. A Theist, of course, can speak much more literally about design, and will likely, on his assumptions, be able to argue more cogent- ly that our cognitive systems have been designed to produce true beliefs. But it is at least conceivable that the evolu- tionary process would also likely produce epistemic systems "designed" to bring about true beliefs because undoubtedly at a certain level of intelligence true beliefs are those which bring about the highest prospects of survival. But for the moment, proper epistemic functioning can be simplistically identified with functioning in accord with one’s epistemic design in and only in appropriate environments. Interestin such a view on it within a the it is highly 11‘ of their functi assume that thi that God desig true beliefs a‘ been created f true beliefs. certain situai sary beliefs. our cognitive istic beliefs tions of othe least attempt neous belief example, dur and intensit 117 Interestingly, the objections which obviously accrue to such a View on a nontheistic basis can also be raised against it within a theistic framework. Granted that assuming Theism it is highly likely our cognitive faculties will have as one of their functions the acquiring of true beliefs, why need we assume that this will always be the case? It is not necessary that God designed our cognitive faculties in such a way that true beliefs are always the intended outcome. They might have been created for other uses which neither require nor produce true beliefs. It may be that they are so devised that in certain situations they give us false but pragmatically neces- sary beliefs. In times of danger, for example, it may be that our cognitive apparatus is fashioned so as to give us unreal- istic beliefs about our abilities, survivability, the inten- tions of others, and so on, in order to allow us to do or at least attempt things we would otherwise not be able to. Erro— neous beliefs are sometimes of great practical benefit. For example, during war it is of immense benefit to fighting moral and intensity to bring your troops to hate the enemy, and this is best accomplished by picturing him as demonic, cruel, and vicious, even when such is not the case. After all, an army which loves its enemy is likely to turn the other cheek, thereby losing the war and, perhaps, much more. A true to life picture of our enemies would paint them too much like ourselves to inspire the sort of intense passion that seems necessary to enable us to kill effectively. There are other possibilities also. Perhaps we have our epistemic apparz than acquiring worship and pra l the accomplishi function, true 1 primary ends. created for th4 and in order t that in certaiI Perhaps God ha to see our n. Fatherly care some random f1 arrive at a f doxastic rect propensity to- sort of divir sary for our MoreoVe] e(iuipment is emotions, fe 118 epistemic apparatus predominately for the sake of ends other than acquiring knowledge. Their employment in the right worship and praise of God, dutiful obedience to His will, and the accomplishing of His purposes might be their intended function, true belief being planned only where it serves those primary ends. Or perhaps as Calvinists believe we have been created for the sole purpose of glorifying and enjoying God, and in order to facilitate that purpose, God so designed us that in certain conditions we come naturally to false beliefs. Perhaps God has done so in order to humble us and so bring us to see our need of Hint and. to appreciate His undeserved Fatherly care and protection. This might be accomplished by some random function which determines that every so often we arrive at a false belief. Thus, perhaps our emphasis upon doxastic rectitude is simply an indication of our natural propensity towards hubris, and so a clear indication that this sort of divine chicanery with our noetic machinery is neces- sary for our spiritual betterment. Moreover, it must also be recognized that our cognitive equipment is not isolated from our other faculties such as our emotions, feelings, and desires. It may be that creatures with our total makeup, placed in an environment such as we are in, simply cannot be designed so as to come to true beliefs in all those situations which are necessarily a part of our ongoing lives. For example, it may be that holding all true beliefs about ourselves or about others would severely hand- icap our ability to relate properly and to have successful a severe misal event. PerhapE original sin v perceive the wo some perverse n ties that we a1 ought to be “i, existence of s The point cognitive facu ed outcome uni of scenarios Theism, but j made. More0i counter-exam}: other, viabl warrant to be faculties ha‘ in appropria- functioning "those sitUa t° bring abc 119 marriages and families. Finally, perhaps our epistemic equipment has undergone a severe misalignment brought about by some catastrophic event. Perhaps Luther and Calvin are right: by virtue of original sin we are "totally depraved" and so unable to perceive the world correctly in certain key respects. Perhaps some perverse moral infection so affects our cognitive abili— ties that we are unable to know certain truths even when they ought to be "properly basic" for us. Indeed, perhaps the existence of such an infection is just such a truth. The point is, even Theism does not guarantee that our cognitive faculties are such that true beliefs are the design- ed outcome under appropriate conditions. Perhaps the sorts of scenarios given above cannot be made consistent with Theism, but if so, an argument to that effect needs to be made. Moreover, even if we are unable to register viable counter—examples, that does not rule out the possibility that other, viable ones lurk nearby. In order for epistemic warrant to be secure, we need an argument that our cognitive faculties have indeed been designed to give us true beliefs in appropriate circumstances. Thus, the criterion of proper functioning in appropriate circumstances must be limited to "those situations in which our cognitive apparatus is designed to bring about true beliefs."12 12Plantinga admits this. See, Plantinga, "Justification and Theism," Faith and Philosophy, Vol. 4 No. 4. October, 1987, p. 414. Consequen‘ God is basic ar to find out wh warranted by w] of proper func but it seems i evidential knc the "proper f1 will simply be certain of cm But this does not insu is properly 1 certain input producing tru we have no r: this basis a2 basic belief because it is BUt let that our bas beliefs are 1 That! hOWeVe] 120 Consequently, in order to know whether or not belief in God is basic and if so rational, we will first need to be able to find out which beliefs, according to the design plan, are warranted by which sets of conditions. Clearly, the criterion of proper functioning is empty unless we have such knowledge, but it seems we will have it only by means of some sort of evidential knowledge. Moreover, unless a blanket approval of the "proper function" criterion can be made, that criterion will simply become a conjunction of individual types by which certain of our prima facie beliefs are warranted. But this entails that the basicality of belief in God does not insure its rationality. For even if belief in God is properly basic because it is the "designed" effect of certain input, unless we know that this design is aimed at producing true beliefs and that, in fact, it is successful, we have no reason to think that God actually does exist on this basis alone. But without such a reason, our properly basic belief in God will not necessarily be rational just because it is basic. But let us suppose it is the case that we are so designed that our basic beliefs are true. We can only know which beliefs are basic if we acquire the design plans themselves. That, however, can only be accomplished by means of authority or empirical investigation. Suppose, on the one hand, we were to accept some set of plans on authority. We would have no way of determining whether or not we actually had the right design plans or if those design plans were aimed at producing . " , 31 pain-013m! . WM! 91.! an!!!» VIM .thaim" days fu-‘fif‘; £11“ 9“ all”. ’1 M' i ' ~ - ‘c 2069:.“ \"d I.'-.-: .1 =. in 310?. Jack ‘ i "5' 'I. " _'-‘-t "'1". Iii-tingbl'. 1 I h I“ I ”an: Ifso,ofwh1 .! l u. I.‘ ' - ' . ' _ . ‘3‘...I .! h I. .l In addi ‘ i a I u whether or , 1 obtained it-- that would I discerning t complish all belief in G evidence an Proper basi. in God woul So P1, helpful in actua11y is it is true as those s 121 true basic beliefs, and if they were, to what extent, unless we were also to have a method of confirming or disconfirming what we were told. But then we would have no need for the authority. On the other hand, if we could determine which beliefs were properly basic by means of our empirical epis- temic systems, we would already have an epistemological method adequate for the discovery and confirmation of true beliefs. If so, of what need are the design plans themselves? In addition, of course, we would have to determine whether or not the design worked, (regardless of how' we obtained it--experimentation, induction, authority, etc.), and that would require the utilization of a procedure capable of discerning true from false beliefs. If indeed we were to ac— complish all this and thereby produce convincing evidence that belief in God is warranted, it would be warranted on that evidence and any warrant accruing to it by virtue of its proper basicality would be unnecessary. In that case, belief in God would be rational even if it were not properly basic. So Plantinga’s View of basicality seems not at all helpful in determining (a) whether or not belief in God actually is properly basic and (b) if it is, whether or not it is true. That is, Plantinga's construal of basic beliefs as those so designed to be does not benefit our search for epistemic foundations in general or for Theism in particular. IV.3 will”: When con: Plantinga h01< beliefs. Whe‘ warrant derive situation, fr which the bel ment between supposedly we by some such ing, one Whit Thus, it make tive faculti tions they 9: ate aspect 1 ranted for : situations. it is clear be aware of MoreOVer, t? it is not experience" Without the \ 130f C‘ §°r juStifv 1“ his acc leave out leaVe the1T 122 IV.3 WARRANTED BELIEF OR WARRANTED BELIEVING When considering the warrant of an evidential belief, Plantinga holds that it is transferred from other warranted beliefs. When it comes to properly basic beliefs, however, warrant derives not from other beliefs, but from the epistemic situation, from the experience in which or as a result of which the belief arises. Thus, there is some sort of entail- ment between the believing situation and the belief which is supposedly warranted. Such entailment can only be justified by some such concept as Plantinga’s idea of proper function— ing, one which presupposes some adequate sense of "design".13 Thus, it makes perfectly good sense to hold that if our cogni- tive faculties have been so designed that in certain condi- tions they give rise to veridical beliefs about some appropri- ate aspect of those conditions, then such beliefs are war- ranted for someone who has those beliefs in the appropriate situations. This is an externalist criterion of warrant, and it is clear that the one having such a belief need not himself be aware of that criterion for his belief to be warranted. Moreover, this criterion is external to the belief itself, for it is not some property of the belief or even the "belief experience" which entails warrant, but rather the "design". Without the proper design, such beliefs would not have warrant 1”Of course, any reliabilist account would be a candidate for justifying this entailment, but Plantinga does seem right in his accusation that insofar as other reliabilist accounts leave out the aspect of "proper functioning", they thereby leave themselves open to damaging counter-examples. such conditior 0n reflec strual of war but rather th« stances exter belief can on If the belief properly basf one could sa for the belie be entailed properly bas unless there circumstance latter like factor, it Warranted _ “ 123 at all. The conditions in which the belief occurs ground the belief, but only because the cognitive apparatus of the believer has been devised in such a way that beliefs had in such conditions by such cognizers are reliable. On reflection, however, it is clear that what such a con- strual of warrant entails is not the warrant of the belief, but rather the warrant of the believing. For how can circum— stances external to a belief warrant the belief itself? The belief can only acquire warrant from other warranted beliefs. If the belief itself had some warranting property such as the properly basic beliefs of classical foundationalism do, then one could say that the experience of the belief was warrant for the belief, for then the truth of the belief would somehow be entailed by the experience itself. But for Plantinga’s properly basic beliefs, the belief itself is not warranted unless there actually is some causal connection between the circumstances of the belief and the belief which makes the latter likely to be true. Without the addition of that factor, it is not the belief, but the believing which is warranted.“ The believing is warranted because the believer “Incidently, this shows that design itself and proper functioning in accord. with design are not sufficient to account for warranted belief. If they were, any type of belief could be warranted by any sort of process, but clearly that is not the case. I might manufacture an automaton so that every time it thought of a prime number it also had a true belief, but surely those beliefs would not thereby be warranted for the automaton. There must be some rational path from the conditions giving rise to a belief to the having of the belief if the latter is to be warranted for a believer. Such a rational connection would undoubtedly be a causal one, i. e. a connection such that conditions of a particular kind when affecting a cognizer of a particular construction will ? conditions of i Plantinga pro] such beliefs lieves in God be warranted ) exists" is we cause its epi (one which re of then). B accord with c i cognizer cou tions are. i its beliefs V as properly shown to be tion. Indee Process of c‘ tific invest problem for is applying conceivable beliefs as under certa belief in G COUld be sh in GOd, but belief! WhE its content cause. Put ( epistemic c Channels c beliefs ar Channel" b‘ belief me< IIarbitrary of above? 124 is so constructed that in such and such circumstances he comes to such and such beliefs. An evidential belief is warranted by those beliefs which. are evidence for it because that evidence entails the truth of the evidential belief. But conditions of belief do not themselves entail the truth of a Plantinga properly basic belief. That we naturally come to such beliefs may be true. Indeed, the 14 year old who be— lieves in God because parents, et. al. so believe, may well be warranted in so believing, but that does not mean "God exists" is warranted too. That belief is only warranted if cause its epistemic systems to produce the appropriate belief (one which represents those conditions or some fitting aspect of them). But if warranted beliefs can only be designed in accord with causal laws, it seems that a sufficiently equipped cognizer could eventually determine what those causal rela— tions are. That is, it could determine under what conditions its beliefs were warranted. But then any belief type designed as properly basic for a specific type of cognizer could be shown to be "properly basic" for a cognizer of that construc- tion. Indeed, this seems to be just what humans are in the process of discovering by means of such things as the scien- tific investigation of perception, memory, and cognition. The problem for Plantinga's position on the basicality of Theism is applying this possibility to belief in God. It is at least conceivable that scientific methods might show that such beliefs as "I am seeing a tree" are "properly basic" for us under certain conditions, but how could they ever show that belief in God is? The most that one can imagine is that it could be shown that in certain conditions we come to believe in God, but that would not demonstrate that the causes of that belief, whatever they turned out to be, were appropriate to its content, i.e., that belief in God had a rational "natural" cause. Put differently, rationality puts constraints upon epistemic design, yi;., the latter must conform to appropriate channels of belief production if the designed cognizer’s beliefs are to be rational. But what would an "appropriate channel" be for Theism? Since the God of Theism is so utterly beyond any physical causal possibility, would not any designed belief mechanism for causing humans to believe in God be "arbitrary", i.e., "outside" the rational constraints spoken of above? God exists an belief, i.e., . In a ser Chisholm. Chi: doing one’s 6 demonstrates , the person’ s i do all I can ' only to the e truth of that ing even if ‘ even if it ha I have at my I may have e believing ir belief is ra person may )3 irrational r the belief : tional act, Consid. Particularl that he Can the fact h physician 125 God exists and/or there is substantial support for that belief, i.e., evidence for its truth. In a sense, Plantinga is in the same situation as Chisholm. Chisholm’s requirement for warranted belief, i.e., doing one's epistemic best, does not, as Plantinga amply demonstrates, warrant a belief. But certainly it warrants the person’s believing. After all, if I have taken care to do all I can to insure that my belief is true and believe it only to the extent to which the evidence I have entails the truth of that belief, then surely I am warranted in so believ- ing even if the belief itself happens to be false, indeed, even if it happens not to be warranted at all by the evidence I have at my disposal. The fact that despite my best efforts I may have erred or malfunctioned does not entail that I am believing irrationally. But if my believing an erroneous belief is rational, surely it is warranted. Contrariwise, a person may believe a perfectly rational belief, but do so for irrational reasons or for no reasons at all. Then, although the belief itself is rational, his believing it is an irra— tional act, an irrational believing. Consider, for example, someone who has been zapped by a particularly rare cosmic ray which so reorganized his brain that he can perform exceptionally difficult surgeries despite the fact he has acquired none of the knowledge a skilled physician would necessarily have about the human body. Indeed, suppose he can not only perform such operations, but does so with precision and guaranteed success. One would want tosaY that “hf queS he uses operations. 1 physicians tra accident gave one would sim knows quantum mechanics are believing; ti rational met? thinking. In short irrationally do the form reasons and bizarre beli which are j) ties.‘5 Plantil one's belie rational in in God is 126 to say that while he may not know why the operational techni- ques he uses work, he certainly knows how to perform the operations. Indeed, he knows how as well as or better than physicians trained in the usual manner. Likewise, if such an accident gave one a precise knowledge of quantum mechanics, one would similarly want to say that the person so affected knows quantum mechanics and that his beliefs about quantum mechanics are warranted. But his believing is not warranted believing; that is, it is not rational, or produced by a rational method, or the result of a rational process of thinking. In short, it is possible for me to hold a rational belief irrationally or to hold an irrational belief rationally. I do the former when I hold any true belief for irrational reasons and the latter whenever I hold grossly false and bizarre beliefs on the basis of beliefs which support them and which are justified for me to the best of my noetic abili- ties.15 Plantinga’s view of proper basicality would show that one’s believing in God was properly basic and, therefore, rational in the appropriate circumstances, but not that belief in God is rational. Of course, this is a problem for any 15Since Plantinga wants to restrict the concept of warrant to those beliefs which not only have justification, but are true, he may want to argue that believings may be justified without being warranted, that only believing true propositions is warranted believing. But then warrant would derive not from other beliefs or appropriate conditions, as Plantinga contends, but from truth. belief produci produces true only by means But then the our determina‘ make those de IV.4 m; In chap cannot be eli some “basic belief-const to be a has would provic‘ their noeti appl‘opriati< be seen tha‘ 127 reliabilist account of basic beliefs. In order for reliably produced beliefs to be rational, they must be appropriately produced true beliefs. But we can only know that a particular belief producing mechanism is appropriate if we know how it produces true beliefs. Clearly, we can, in turn, know this only by means of producing evidence that this is the case. But then the reliability of the mechanism is superfluous to our determination of justified and warranted beliefs. We can make those determinations by other means and on other bases. IV.4 IMIEBHALl§I_QQH§EBAlHI§_AflD_IXEE:EQKEH_DLEEIQHLIIE§ In chapter VI, we will see that internalist concerns cannot be eliminated from epistemic warrant, that for at least some "basic beliefs" to be warranted, certain internalist belief-constraints must be met. For instance, for a belief to be a basic belief, one must not have any beliefs which would provide reason for rejecting it and one must have in their noetic structure beliefs sufficient for the proper appropriation of a basic belief. In addition, it can easily be seen that the need for evidential arguments in support of a Plantinga properly basic belief also arises from internalist constraints. The need for defeater—defeaters, for example, derives from the fact that Plantinga basic beliefs are defeas- ible, and if one is in possession of a belief or set of beliefs which undermines a "basic belief", then that belief or' belief set. must be addressed. But again, it is the internalist aspects of Plantinga’s view of warrant that determination for Plantinga will be impOSS type which i] occurrence of But note tha‘ myriad of typ sified under which one is Even we becomes a su] marshalled i view is a c show below, "PrOper func internal n01 128 creates that demand. These internalist.constraints present.another’difficulty. Given the fact that they must be met for each belief, the determination of the correct warrant rule (the R of our schema for Plantinga properly basic beliefs) for a particular belief will be impossible to determine. Each R will specify a belief type which in conditions C will be warranted, while each occurrence of a belief so warranted will be a belief token. But note that every particular "basic belief" will have a myriad of type possibilities. Since each token can be clas- sified under a number of different types, how do we determine which one is appropriate, which one is legitimately warranted? Even were externalist concerns alone considered, this becomes a supremely difficult problem. Indeed, Plantinga has marshalled it against Goldman’s View.16 But while Goldman’s view is a completely externalist one, Plantinga's, we will show below, has internalist elements due to his inclusion of "proper function" within his criterion for warrant. Thus, the internal noetic structure must be such that it is appropriate for a particular basic belief, otherwise it will not be able to function as designed in a particular epistemic situation, and, consequently, warrant will not accrue to that "basic belief". 16See Alvin Plantinga, "Epistemic Justification," Nous, 20 (1986), pp. 8—12, and “Positive Epistemic Status and Proper Function," Philosophical Perspectives: Vol. II, Epistemology, James E. Tomberlin, ed. (Atascadero, California: Ridgeview, 1988), 1—50. Indeed. 1 every Particul warrant. 311“ belief" will belief types: we cannot est: ion of proper here, for the specificatior tions is appl In addi' met, the lisi they will ha with the pr< function do Thus, present 111 epistemic : belongs to belief" typ can be det belief", f C0nditions ditions sh by each R determine Whether C 129 Indeed, it is possible that every particular belief in every particular situation will require a specific R for its warrant. But regardless, it is clear that every "basic belief" will be classifiable under" a number of different belief types, and unless we know which one is the correct one, we cannot establish its warrant. Notice too that the criter- ion of proper function according to design does not help us here, for the problem is not in not knowing what the design specifications are, but in determining which of the specifica- tions is appropriate for warranting a particular belief. In addition, given that internalist constraints must be met, the list of conditions C for each R will be infinite, for they will have to rule out every belief which would conflict with the proposed "basic belief". Again, design and proper function do not eliminate this problem. Thus, the main problem these type-token difficulties present lies in the resulting ambiguity regarding their epistemic status. If a particular "basic belief" token belongs to a number of different warranting rules ("basic belief" types), which one is the appropriate one? Unless that can be determined, one cannot assess its status as a "basic belief", for one cannot determine whether or not the proper conditions have been met unless it is clear what those con— ditions should be. Moreover, if the list of conditions given by each R is indeed potentially infinite, even if we could determine which R is the proper one, we could not determine whether or not its conditions were satisfied. But then we have merely PC belief—worthy of course, on nevertheless L know that to that any part IV.5 w A furthe cality as it determining v belief in G questions ar< sorts of conc‘ we cannot de if we know t What sorts o: it is so, C Possible to are PrOperlj 130 have merely potential basic beliefs, not beliefs known to be belief-worthy by virtue of the conditions giving rise to them. Of course, on a purely externalist account, many beliefs might nevertheless instantiate "basic beliefs", but we could never know that to be the case, and, therefore, could never know that any particular belief in God was "basic". IV.5 THE JUSTIFICATION OF THEISM’S BASICALITY A further problem with Plantinga’s View on proper basi- cality as it relates to Theism involves the difficulty of determining whether or not and, if so, under what conditions belief in God. would. be properly’ basic. Indeed, the two questions are actually one, for unless we can know under what sorts of conditions belief in God is Plantinga properly basic, we cannot decide if it should be so considered. Conversely, if we know that it is Plantinga properly basic, we must know what sorts of conditions or believing circumstances exist when it is so. Certainly, for other Plantinga basic beliefs it is possible to outline the sorts of conditions under which they are properly basic. For example, I know I am justified in believing basically that I see a tree if I am appeared—to- treely and other conditions are such that it is unlikely I am hallucinating, being deceived, etc. With some effort, we can at least make a start in listing just what sorts of conditions actually need to be the case for my belief that I see a tree to be warranted. But what sort of conditions would be the analogue for disanalogies I 1y basic beli other observ: objects are 1 appears to be common objec Such se regard to be are possible objects two same object taneously, to God in s that object Given particular descriptio be metaphe being, Ne in terms Indeed, 6 Vision wy Theism Gc being Se 131 belief in God? Unless those can be specified in at least general terms, we have no reason to believe belief in God actually is Plantinga properly basic. It is here that the disanalogies between belief in God and other Plantinga proper- ly basic beliefs becomes clear. For example, in regard to other observation beliefs, valid observations of physical objects are learned by ostensive definition. At least that appears to be how we learn them. For such beliefs there are common objects which different observers can point to. Such seems clearly not to be the case, however, with regard to belief in God. Even if direct experiences of God are possible, there is no way to determine whether or not the objects two observers of God have experienced are actually the same object; indeed, even if they are observing God simul- taneously, neither we nor they could tell. One cannot point to God in such a way that the other will see him pointing to that object. Given the impossibility of observationally confirming a particular experience as one of God, one might think verbal descriptions would be sufficient. But such descriptions would be metaphorical, for the God of Theism is not a physical being. Nevertheless, those descriptions will of necessity be in terms of words which refer to the physical universe. Indeed, even if there were some mutually viewed external vision which was an appearance of God, since according to Theism God is a spirit, it would not be God Himself who was being seen, but some "makeshift" appearance designed to indicate God's Thus, we hand, there is private, innel than our knov other hand, 6 or experience beyond the da object of the do we have ar experience a establish an‘ i.e., we car So even if belief, no < The are in God tak arguments a belief or a f iculties v 132 indicate God’s spiritual presence.17 Thus, we are presented with a dilemma. If, on the one hand, there is direct experience of God, it is necessarily a private, inner experience which utilizes some faculty other than our known epistemic observation systems. If, on the other hand, experience of God is indirect, by way of objects or experiences :hi the physical world, the inference goes beyond the data, for no physical experience entails that the object of that experience is God. Moreover, in neither case do we have any means of providing criteria of identity for an experience as an experience of God. But then we cannot establish any conditions for the proper basicality of Theism, i.e., we cannot say when belief in God is or is not basic. So even if it happened, to be a ZPlantinga properly' basic belief, no one (except God of course) could know it. The argument of the last few paragraphs applies to belief in God taken as a basic observation belief, but similar arguments apply if one considered it some sort of innate basic belief or an authority basic belief. If anything, the dif— ficulties would then be even more severe. 17An orthodox Christian might object here that Christ is God in the flesh, the fullness of God in bodily form. Even so, a vision of the resurrected Christ is not an unambiguous vision of God Himself. Christ may indeed be God incarnate, but the being of God itself is not thereby seen. Rather, one sees the human body of the incarnate God. (3) (4) (5) 133 IV-6 QQHQLHfiIQfl To summarize, the following have been argued: (1) No matter what type of Plantinga properly basic belief belief in God is taken as, it is not strongly basic enough to remain such given the awareness of possible defeaters. It then becomes, in one sense or another, an evidence essential belief. This is due to the relative weakness of Plantinga basic beliefs and indicates a basic weakness in "weak foun— dationalism". (2) Plantinga’s inclusion of "functioning in accord with epistemic design" as part of the criterion for warranted belief makes it imperative that we know the design plan before we can determine which beliefs are properly basic and under what conditions. But if we can accomplish that, we already have a method of determining true and false beliefs, and do not need to know which beliefs are and which are not properly basic. (3) Plantinga’s position actually provides for the warranting not of beliefs, but of one’s believing. It is the former, however, which is needed if Theism is to be rational. This weakness is due to externalism in general. (4) The internalist constraints which result from the “proper functioning" clause create type- token and other warrant rule identification difficulties which undermine the epistemic value of Plantinga basic beliefs. Such difficulties seenlto be endemic to reliabilism. (5) There are severe difficulties in determining whether or not and, if so, under what condi- tions belief in God is properly basic. Regard— less, if it is, Theism’s epistemic status is not abetted unless we know that it is, but if we can determine that, we no longer need the epistemic value such a fact would otherwise provide. If the arguments supporting these conclusions are sound, we can conclude that the epistemic status of Theism as a Plantinga properly basic belief is not a significant improve— ment over that which it has as an evidence essential belief. . to me .m‘lione no same. saw at maul-II semifinals: art: u: sub 31 ain't nae-1i 13d {sumac rm.- zisrlsd 2:23:36. -r;=.~si_..! .wi'. 39 33:1".an -'-':.-:':"-‘ #24110“ :11 anemia-9' ”.2996 s- JESSJ'hflI. - ' :' -"'- ' J.‘ 2‘.- -.- _ v' s '- i for belief in that Theism if 134 This is so because as a Plantinga basic belief it does not itself have warrant unless it can be shown not only that it is basic for us, but also that such basicality entails its truth. The demonstration of those facts, however, will lend Theism as much or more rationality than it would have as a Plantinga basic belief. Indeed, if it can be shown that belief in God is properly basic, we therewith have evidence for belief in God. It is, therefore, as an evidential belief that Theism will flourish most epistemically. Planting category of l argument thai referentiallj presents for His example the teaching circumstance need not be this as juS‘ PrOperly bas 0f beliefs basic in ti (hereafter not only be CHAPTER V AUTHORITY BELIEFSl Plantinga’s case that CF has too narrowly delimited the category of properly basic beliefs is based not only on his argument that the CF criterion for proper basicality is self- referentially incoherent, but also on the arguments he presents for the proper basicality of other sorts of belief. His example of the 14 year old who believes in God because of the teaching of parents, etc. is a case in point. In certain circumstances, he holds, such a belief is properly basic and need not be grounded in evidence. Although Plantinga uses this as just one example relevant to just one sort of nonCF properly basic belief,2it is a key example of a certain class of beliefs which, if Plantinga is correct, will be properly basic in the right circumstances, yi;., authority beliefs (hereafter abreviated ABS). In certain situations, they will not only be rational, but rational without being grounded in 1By "authority beliefs" I mean "beliefs based on author- ity". We shall use the former phrase throughout as equivalent to the latter, thereby eliminating what soon becomes cumber- some and awkward phrasing. Indeed, in deference to our current penchant for ontological parsimony, we shall usually abreviate to a sparce "ABS". 2As we have seen, observation and memory beliefs are two other types of beliefs which Plantinga holds are properly basic in the right circumstances. 135 any evidence. 1 have a beliefs-—incl' Plantinga's foundationali maximally ceI further subsi argued contr; in God is has ing of some will first valid, for a which do no‘ are themsel based on be Second, I w beliefs are rational no results, t] allowed by Secti‘ holding th properly 13 right in a be Claimed tion Ollghi attempt in 136 any evidence. I have argued, however, that Plantinga properly basic beliefs-—including authority beliefs—-are not, contrary to Plantinga's contention, properly basic in any properly foundationalist sense of that term, for they are neither maximally certain nor such that they are without any need of further substantiation (Chapter III). Consequently, I then argued contrary to Plantinga’s View, that insofar as belief in God is based on authority, it will need evidential ground— ing of some sort (Chapter IV). In this chapter, however, I will first contend. that Plantinga’s criticisms of CF are valid, for authority beliefs are a class of rational beliefs which do not satisfy CF criteria for rationality since they are themselves neither CF properly basic nor evidentially based on beliefs traceable soley to properly basic beliefs. Second, I will also concur with Plantinga’s View that such beliefs are nevertheless both rational and essential to our rational noetic structures. Consequently, given our previous results, they must be grounded in some way other than that allowed by either CF or Plantinga. Section V.1 argues that although Plantinga is correct in holding that some ABs need not be evidentially grounded in properly basic beliefs to be rational, he is nevertheless not right in ascribing to them proper basicality. ABS, it will be claimed, do have justification and without that justifica- tion ought not be accepted as rational. Section v.2 will then attempt to show that authority beliefs are necessary elements in a person’s grounded in Pr CF. Finally, theless, warr authority be] neither prope be a third ty that of eithe V-l Aim; In this not properly Plantinga’s Theist does basic in an: supported b Planti in God bec; trust so b 137 in a person’s rational noetic structure and yet cannot be grounded in properly basic beliefs in a manner consistent with CF. Finally, section v.3 contends that most ABS are, never- theless, warranted beliefs. The conclusion reached is that authority beliefs are rational despite the fact they are neither properly basic nor CF evidential. Hence, they must be a third type of belief the grounding of which differs from that of either basic or evidential beliefs. V-l AHIflQBlI1_BELlEE§_AHD_BBQBEB_§A§l§ALlIX In this section, I shall argue that authority beliefs are not properly basic beliefs. It will be contended (a) that Plantinga’s argument from the example of the 14 year old Theist does not entail that authority beliefs are properly basic in any sense and (b) that authority beliefs are in fact supported by nonCF evidence. Plantinga’s example of a 14 (or 8) year old boy believing in God because all those around him whom he has learned to trust so believe is a case for which some foundationalists might agree that evidentialist sorts of support are unneces— sary. But what of Plantinga’s example itself? Surely by itself it does not undercut (CF) evidentialism for rational adults, for one need not maintain that the criteria for belief applicable to adults holds for children also. One could simply add the reasonable amendment that the evidence require~ ment for nonbasic beliefs applies only to adults. More importantly, this example does not support an}: mm an 91: um 3931 ed: 931ml: Inna-13:1 sac 31.1.1.6 ..‘eu-t. --":"i'." .P‘urptaafi .23.i:*r'-e.zé;.'~r-:- --.2- can tit”?! ':I'nq01§ "i " .;.--.- ' -:'c 'E .'..-".\-: "- - p"? - '- -: "- .'--:Z ' erg-:3 {flirts I” d . .- H*l' 5C 1nd? this claim is example, eve; hold.) But contrary to stitutes fur basicality follows: (1) Ac p1 p: (2) At 11 (3) A b: bt (4) B (5) B b: (6) T] i (7) TI wi Obviou 138 Plantinga's main contention either. He claims, correctly I shall argue, that this is an example of a belief which neither satisfies the CF criterion for proper basicality nor is derived evidentially from properly basic beliefs and yet is rational for the believer. (Since my argument in behalf of this claim is not dependent upon the use of a child as is this example, even were we talking about a rational adult, it would hold.) But he concludes, incorrectly I shall contend, that contrary to CF, it is pnoperly basic, and therefore con— stitutes further evidence that the CF criterion for proper basicality is faulty. His argument can be outlined as follows: (1) According to CF, rational beliefs are either properly basic or evidentially derived from properly basic beliefs. (2) According to CF, a belief is properly basic iff it is incorrigible or self—evident. (3) A 14 year old who accepts a belief B taught him by parents, teachers, etc. is rational in so believing. (4) B is not CF properly basic. (5) B is not derived (derivable) from properly basic beliefs. (6) Therefore, B is a properly basic belief which is not CF properly basic. (7) Therefore, there are properly basic beliefs which are not CF properly basic. Obviously, however, conclusions other than (6) and (7) are derivable from (1)-(5). To draw (6) and (7) one needs 2 other premises, (5a) Some form of foundationalism is true, and (5b) All forms of foundationalism permit only 2 categories of rational propositions: properly basic and those evidentially'derived'therefrom. (I.e., (1) holds for all forms of foundational- ism.) In addition, ‘ (5). Let us f used this art why not inst' draw (6b)? (6b) So In other wor rational be] beliefs can not then be beliefs are those old e In ord correct om permises (I Properly b proper pro sive argun doubtless are not CF I Shall ar are not C‘. 139 In addition, it is certainly possible to question any of (l)- (5). Let us for the moment accept (4) and (5). Plantinga has used this argument to reject the CF view stated in (2), but why not instead reject the CF position contained in (1) and draw (6b)? (6b) Some rational beliefs are neither properly basic nor evidentially derived from properly basic beliefs. In other words, why not conclude either that other classes of rational belief exist besides the 2 recognized by CF or that beliefs can be supported by other than CF evidence. One would not then be forced to agree that this example shows that such beliefs are properly basic, not even for children—-at least those old enough to reason.3 In order to show that this alternative conclusion is the correct one to draw, it will be necessary to show (1) that permises (3)—(5) are true and (2) either (a) that ABs are not properly basic or (b) that Plantinga basic beliefs are not proper properly basic beliefs. We have already given exten- sive arguments for (2b) in chapter III. Premise (4) would doubtless be admitted by everyone; clearly authority beliefs are not CF properly basic. In the remainder of this section, I shall argue in support of Plantinga’s view that such beliefs are not CF evidential beliefs (for premise (5)), but against 3These are the only one’s we are interested in, for rational belief is under consideration, and those who cannot reason are nonrational. Moreover, self—conscious rationality is important. his contentit Let us I er. Certaini about his be ates, form t My parents a bushel Therefore b1 author: Beliefs baSI belief Therefore a Rational be Therefore r M Therefore I Consequentl to the be] argument. we would n fact, a ca disbelief More as well, ; that an a( authority lst Crite 140 his contention that they are basic (for (2a) above).4 Let us look more carefully at such a 14 year old believ- er. Certainly such a person does not reason (CF) evidentially about his belief. He does not, as Plantinga rightly insinu- ates, form the syllogistic sortie: My parents are proven trustworthy authorities. My parents believe in God Therefore belief in God is based on proven trustworthy authorities. Beliefs based on proven trustworthy authorities are rational Qaligigl Therefore a rational belief is belief in God. Rational beliefs ought to be believed by rational agents. Therefore rational agents ought to believe in God I am a rational agent. Therefore I ought to believe in God. Consequently, he certainly does not decide to believe or come to the belief that God exists on the basis of the above argument. Moreover, Plantinga is doubtless right in holding we would not call him either irrational or nonrational. In fact, a case might well be made for the irrationality of his disbelief in God in that particular situation. More importantly, similar situations exist with adults as well, i.e., situations wherein one would not want to say that an adult was irrational although she believed some p on authority which did not, for her, meet classical foundational- ist criteria. In our everyday lives we frequently hold ‘The next section, then, will argue both that authority beliefs cannot be CF justified and that they are ineliminable from our noetic structure, while section V.3 will present further argumentation in behalf of the rationality of author- ity beliefs. This will then lead us to conclude that rather than reject (2) as Plantinga does, one ought to reject (1) and allow for beliefs which are rational despite the fact they are neither properly basic nor CF evidential. beliefs on th authority ly scientific ’61 nature and m not, in mam which it ma! argument whi which we ne‘ Am I, hibernate, is the res Kellogg’s t ies, that Aramaic? 1 constructe‘ believed t trustworth all the ab impossible create a above, ar rationali pert's--f But 1y deriv. for him WhiCh SU 141 beliefs on the authority of others, indeed often on chains of authority lying behind the immediate authorities. Current scientific theories, news events, and a host of "facts" about nature and natural processes are things (1) for which we do not, in many cases could not, have direct evidence, (2) for which it may or may not be possible for us to construct an argument which would satisfy foundationalist scruples, and (3) which we nevertheless believe rationally. Am I, for example, irrational in believing that bears hibernate, that the killer bees infestation in Latin America is the result of a Brazilian experiment gone awry, that Kellogg’s bran flakes in the Variety Pack size have 70 calor- ies, that Latin for "bronze" is "ass", or that Jesus spoke Aramaic? All of these are beliefs I have without ever having constructed an evidentialist argument in their behalf. I have believed them on the "word" of others whom I believed to be trustworthy and knowledgeable. Am I irrational in believing all the above? Surely not. It may even be difficult, perhaps impossible (some of them might be false), for the experts to create a foundationalist chain of evidence for some of the above, and. yet it. would. be reasonable to assume that ‘my rationality would nevertheless remain intact, as would the ex- pert's—-for now at least. But while the 14 year old's belief may not be evidential- ly derived from properly basic beliefs, neither is it basic for him in the sense that he does not have available data which substantiates or supports his belief in the authority and the meal terns which Church, etc. for him. G] of this rat probably co support doe cates that Clearly, t basicality were justi at all, hi epistemica In fa less rigic for such r namely, t appear to of being God, a Ch "Because His belil beCGUSe . 142 of those whom he is believing. His "believing situation" acts as an (at least) implicit rational support for his belief. Thus, his past experience with those authority figures and the consequent beliefs he has about them, the contexts in which and the means by which he is so taught, the behavioral pat— terns which he has learned (praying at bedtime, attending Church, etc.) all play a part in "substantiating" this belief for him. Granted that for children the conscious utilization of this rational support does not and, for the vast majority, probably could not be formalized by them, the fact that such support does exist and is implicitly utilized by them indi- cates that such beliefs are not properly basic themselves. Clearly, they do not satisfy' the requirements for' proper basicality set out in the Chapter III, for even if the child were justified in so believing in God without any CF evidence at all, his belief would not be either maximally certain nor epistemically final for him. In fact, these beliefs do not even appear to satisfy the less rigid view of proper basicality which Plantinga accepts, for such beliefs seem plainly supported by justificatory data, namely, the authority of parents, teachers, etc., and these appear to play a more "active" role in his belief than that of being mere “conditions" for it. Asked why he believed in God, a child in the situation described would doubtless reply, "Because my parents told me," or something to that effect. His belief, that is, is not basic even for him, but believed because of other beliefs, viz., the authority of his parents, etc. AS 3139‘ by him as CF deduce or in this belief belief, a be not held wi1 the authori‘ belief), bu‘ in virtue 0 Part 0 he has of themselves or to be sc them on th were the c alone. So him, but I beliefs wl But 1 that this Support h 140) in o fOUndati( ground u' Warrant, which ar beliefs. 143 etc. As argued above, such supporting beliefs are not taken by him as CF evidence for belief in God; he certainly does not deduce or induce belief in God from them. Rather, he holds this belief on authority. Therefore, it is an authority belief, a belief held neither basically (i.e., the belief is not held without any belief support) nor evidentially (i.e., the authority is not taken as evidence for the truth of the belief), but simply by means of or on account of the authority in virtue of which he holds the belief. Part of what justifies his belief is background knowledge he has of his parents, relatives, etc. Had they all shown themselves to be either patently wrong on most other matters or to be scandalous liars, he would have good reason to doubt them on this issue and probably would. Certainly, if such were the case, he ought not believe them on their authority alone. So it seems that this belief is not properly basic for him, but rather is accepted by virtue of a host of background beliefs which give it warrant. But have we not then contradicted our earlier assertion that this young believer does not and need not evidentially support his belief by means of the sortie outlined above (p. 140) in order to qualify as a rational believer? Only if our foundationalism requires that justification come from the ground up. Another possibility also exists, however. His warrant, I maintain, is not linearly derived from beliefs which are more foundational, until one reaches properly basic beliefs. Rather, it is derived from nonproperly basic beliefs which obtain beliefs. HC beliefs d0 n4 to basic beli they support on the “same standard: b‘ Nevertheless Similarly, ' mally certa the 14 year he believes he uses at properly be While that authc would not derivable rationally them from that they beliefs m1 alone if 144 which obtain their warrant from other horizontally confirming beliefs. Horizontally, as opposed to vertically, related beliefs do not lie further down a justificatory path, closer to basic beliefs, but are "on the same level" with the beliefs they support. For instance, my belief that I see a tree is on the "same level" as my belief that my viewing situation is standard; both are observations about the external world. Nevertheless, I might use the one in support of the other. Similarly, while he does not trace his belief back to maxi— mally certain beliefs which require no further justification, the 14 year old does utilize a number of other beliefs when he believes in God because he has been taught to. Primarily, he uses authority beliefs, and these beliefs are neither properly basic nor, it will be argued, derivable therefrom. While a classical foundationalist would certainly agree that authority beliefs are clearly not properly basic, he would not so quickly accede to the claim that they are not derivable from properly basic beliefs. The fact that a person rationally believes some things on authority without deriving them from properly basic beliefs alone surely does not entail that they cannot be so derived. One could argue that such beliefs must be potentially derivable from properly basic ones alone if they are to be rationally held by anyone, although the believer himself need not perform the derivation. The de facto believing without CF evidence of a CF evidence essential belief, it could be claimed, receives its warrant from the fact that (a) such evidence actually is available, (b) the requisite del er has perfol particular 6 despite appe simply are r CF should bl be necessary ed from our carried out (c) ABs are demonstrate (c). v.2 m It is constitute we use a Plantinga in the rig maintain finality , if they a Stemicall in fact n bility. neither be that 145 requisite derivation could be carried out, and (c) the believ- er has performed all the epistemic obligations he has in this particular epistemic situation. Or, it could be contended, despite appearances and our arguments above, such beliefs simply are not rational. Thus, in order to demonstrate that CF should but cannot adequately handle such beliefs, it will be necessary to show that (a) such beliefs cannot be eliminat- ed from our noetic structure, (b) CF derivations cannot be carried out for at least some noetically essential ABs, and (c) ABs are rational beliefs. The next section attempts to demonstrate (a) and (b), and the section following argues for (c). V.2 AUTHORITY BELIEFS AND CF JUSTIFICATION It is not only clear that we do have ABs and that they constitute a significant number of our beliefs, but also that we use a large number of then! to justify other" beliefs. Plantinga has argued that authority beliefs are properly basic in the right circumstances, but as we have seen, in order to maintain this, he cannot ascribe to them the certainty or finality we have maintained properly basic beliefs must have if they are to be properly basic in any meaningful and epi— stemically significant sense. Moreover, I have argued that in fact nonCF evidence is relevant to their rational accepta- bility. Surely, they are not CF properly basic, for they are neither incorrigible nor self-evident. While it might well be that "it goes without saying" that Einstein was an W1 authority 0“ authority 0f is belief in ible; many F points of ph incorrigible beliefs lac‘ ought to c] authority bi basic belie A CF j be justifie so, two t} First, any 1),, which h CF justifi an author asserting ting Pr- standards hold that cation f0 either th (4) be a; tion tha‘ position I only I 146 authority on physics, it is not self-evident. Nor is the authority of any other authority self—evident. But neither is belief in the authority of Einstein in physics incorrig— ible; many physicists of his day disagreed with him on some points of physics, but surely if belief in his authority were incorrigible, they would not have. Likewise, authority beliefs lack the certainty and finality which I have held ought to characterize properly basic beliefs. So either authority beliefs can be CF evidentially based on CF properly basic beliefs or they are legitimate for some other reason. A CF justification of them will need to show how they can be justified on the basis of properly basic beliefs. To do so, two types of justification will have to be provided. First, any authority A will need CF justification for any AB, p“ which he believes. Second, a nonexpert believer will need CF justification for at least the following: (1) that A is an authority, (2) that A is asserting p“ (3) that A is asserting p5 sincerely, and (4) that A is justified in asser— ting pA. With respect to this last condition, two possible standards can then be set for ABs. The strong standard would hold that (4) would entail the nonexpert knowing the justifi- cation for p,which A.knows. The moderate standard could hold either that (4) be equivalent to (l) or, less moderately, that (4) be applied directly to a p“ i.e., that there be confirma- tion that A is justified in asserting p“ Let us call these positions 5*, M-, and M+ respectively. Then, according to M- I only need to know that Peters is a reputable physicist to rationally 13' whereas acCO right about If one beliefs wouI although th Ideally, eve which did n Given' of a ratior question wh sity or mer ity would believers, ality wit] reasons, s they woul believers they woul tions). temologic rational 147 rationally believe what he says about electromagnetic waves, whereas according to M+ I would need corroboration that he is right about this particular matter. If one were to hold to the strong thesis, authority beliefs would be logically or epistemologically eliminable, although they might neverthelesss be practically useful. Ideally, every believer would have a rational noetic structure which did not include any ABs essentially. Given the moderate thesis, ABs would be a legitimate part of a rational noetic structure, but it would remain an open question whether their presence was an epistemological neces- sity or merely a practical one. If the latter, the possibil- ity would exist that at some future date for some future believers, ideal rationality could be attained (i.e., ration- ality without ABs). Although others who, for practical reasons, still needed to rely on ABs would still be rational, they would not be as rational as these projected future believers since they still relied on and held ABs (i.e., since they would not know the CF justification for some proposi— tions). On the other hand, if there is some sort of epis— temological necessity for humans (and perhaps all finite rational beings) to utilize ABs, then rationality will not require that a CF justification for ABs exist. We shall argue that there is an epistemological necessity for ABs for humans, (and likely for all finite rational beings), and that no CF justification can be given for all of them. That is, we shall argue that there exists for each rational being a set of ABs which are ess To show position mor since it is 1 it will be b that the "ba belief-wort one or more often than other belii irreflexive the evident RBG, 52-51 coherentis noetic str belief B structure such that (RBG, 53) basis", ( level in j of A is ] aSSigns 1 148 which are essential members of his rational noetic structure. To show this, we will first lay out the foundationalist position more precisely. Plantinga has done this in RBG, and since it is his criticism of foundationalism we are examining, it will be best to use his explication of it. Plantinga notes that the "basis relation" (that relation which establishes the belief-worthiness of one belief on the belief-worthiness of one or more other beliefs) is a "one-many" relation, for more often than not one belief will be "based on" more than one other belief (RGB, 52). Moreover, this relation is both irreflexive ("self-evident" propositions are not believed "on the evidence of themselves" but rather "without evidence" (see RBG, 52-53)),5 and asymmetric (foundationalists are not coherentists). He then defines more precisely the "rational noetic structure" of foundationalism. "Let us say that a belief B is an immediate basis of a belief A in a noetic structure N if A is based on B in N and there is no belief C such that C is based on B in N and A is based on C in N." (RBG, 53) Then (1) "Every nonbasic belief has an immediate basis", (2) "Every belief B in N will belong to a highest level in N", and (3) "If A is based on B in N, then the level of A is higher than the level of B." (RBG, 53) Plantinga assigns basic beliefs to the "0th" level. Given (2) each 5Chisholm notes the ambiguity in the name and says that one can think of these propositions as analogous either to the unmoved mover who does not move or to the unmoved mover who moves himself. See Chisholm, Theory of Knowledge, 2nd ed., p. 25.) belief will 3‘ (3) the basi Moreover, fr it follows 1 “path" from belief will a belief or major conse that all e traceable t Where, structure? basic neit] down previ justificat arguments. will be Cl tion for < AB would 149 belief will be assigned to one and only one level, and given (3) the basis relation is both asymmetric and irreflexive. Moreover, from these and the foundationalist noetic structure, it follows that for every rational belief there should be a "path" from that belief to a basic belief in which path each belief will be based on an immediately preceding belief (i.e., a belief one level lower on the path). (RBG, 52-54) The major consequence of this position, it should be noted, is that all evidential beliefs must have one or' more paths traceable to properly basic beliefs alone. Where, then, would authority beliefs fall in this noetic structure? As we have already made clear, they are properly basic neither in the sense defined by CF nor that which I laid down previously. To hold that they are in no need of further justification would simply be to legitimize ad verecundiam arguments.‘ Therefore, our belief in the authority of others will be CF rational only if we can construct a CF justifica— tion for our belief in their authority. In other words, each AB would have to have one or more paths each of which ended in properly basic beliefs alone. Moreover, every belief in each of these paths would similarly need to have a path or paths traceable to properly basic beliefs alone, and so on. It is doubtful that anyone, including most philosophers, could 6To put it another way, as any basic logic text notes, arguments from authority are only as good as the substantia- tion of that authority. As outlined in more detail above, this substantiation must be of two different sorts: (a) the substantiation that S is indeed an authority on X and (b) S’s ability to substantiate his claims about X. do that for if our autho structure, S justificatic to CF justi rely on beli manner and v on other au‘ be necessar (i.e., that all authorf we must de from human rational n authority Here thesis. I beliefs re rational 150 do that for most all their authority beliefs. In addition, if our authority beliefs are endemic to our rational noetic structure, some or all of our rational beliefs will lack CF justification. We will try to demonstrate that in any attempt to CF justify one’s authority beliefs, one must eventually rely on beliefs which one has not received in a CF legitimate manner and which one cannot now justify without again relying on other authority beliefs. To do this convincingly, it will be necessary to show (a) that authority beliefs are pervasive (i.e., that they infect all of our beliefs) and (b) that not all authority beliefs can be CF justified. In other words, we must demonstrate that authority beliefs are ineliminable from human rational noetic structures and that one's total rational noetic structure requires some of these ineliminable authority beliefs. Here one can distinguish between a strong and a weak thesis. The weak thesis holds only that some of our rational beliefs require ineliminable authority beliefs, but that other rational beliefs can be jUStified without relying upon them in an essential way. The strong thesis holds that none of our rational beliefs are free from essential infection by author- ity beliefs—-that all of them at some point in their complete rational justification must utilize authority beliefs. My intuition is that the strong thesis is correct, but given the difficulty of proving that without omniscience, I shall argue only for the weak thesis. Take, for example, my belief that helium expands when heated. I C finding the that would 1 direct waY “ suitable com the pressur procedure a: on authorit helium, the that increa beliefs I complete 3" would have Moreo paths alsc of author justify m} I would h me it was a college involve 8 identity learning beliefs, Properly behalf, 151 heated. I could, of course, try to confirm this belief by finding the appropriate sentences in a chemistry book, but that would involve a plethora of authority beliefs. A more direct way would be simply to get some helium, place it in a suitable container with some sort of meter which would measure the pressure inside, and heat it. But even such a simple procedure as this would involve taking a number of key beliefs on authority. For example, that the gas I procurred was helium, that the meter measured the presure of the gas, and that increased presure indicates expansion of a gas are all beliefs I would have on the authority of others, and any complete justificatory path of "helium expands when heated" would have to include them. Moreover, it would have to include their justification paths also, and such paths would simply multiply the number of authority beliefs employed. For instance, in order to justify my belief that the gas I was using was indeed helium, I would have to verify the authority of the person who told me it was, perhaps a bottled gas salesman or a professor at a college. A verification of this person’s authority would involve a verification of his credentials, his integrity, his identity, and his reliability, all of which would involve learning new information and relying upon other authority beliefs. But since those authority beliefs would not be properly basic, evidence would have to be gathered in their behalf, all of which would have to be traceable to properly basic beliefs alone. But surely, this would not only involve a .1 :1 song. .wmuemu an a m Huaaam blyow n‘ureiw 19.199! to :Hos some new all r... -.' 4, .3 F'f. .Gbifin: '-l.¢l.3. '_= " -' L's "fr!“ II.I_‘. - - .- ... e.. '.'. support nece of beliefs 3n justify an 1 progressive justificati process. exist some since each have a ju again, mos on.’ Pro] some CF b 152 learning new information, but also relying upon still more authority beliefs. It can easily be seen that the problem has no plausible solution, for while CF justification demands the progressive narrowing of justifying beliefs to a set of basic beliefs, the CF requirement that all nonbasic beliefs be given evidential support necessarily creates the actual expansion of the number of beliefs which need justification. Any rigorous attempt to justify an evidential belief in accord with CF standards will progressively increase the number of beliefs requiring CF justification, thus making it impossible to ever complete the process. So in order to justify some belief B, there must exist some path V to B which contains evidence Er . . EV But since each E1 also needs to be justified, each one will also have a justificatory path. V1 which. will contain beliefs, again, most of which will be in need of justification, and so 7 on. Probably some of these justificatory paths will have some CF basic beliefs, but given their narrow range and the 7This, of course, is assuming that the justification of an Eirequires beliefs in addition to En“ While occasionally there are immediate inferences from one belief to another, in empirical arguments this is not usually the case, so most all beliefs in such arguments require more than one belief for their support. difficulty 0 justificator Moreover, as its justifi occasionallj tory paths, belief for empirical I justify eve tificatory the furthe quirements one anothe It sh are only i justificat 153 difficulty of justifying anything by such beliefs alone,8 no justificatory path will consist soley in CF basic beliefs. Moreover, as long as each Ei requires at least 2 beliefs for its justification, the number of IAS will increase. While occasionally immediate inferences are contained in justifica-' tory paths, most often a belief requires more than one other belief for its justification. This is especially-true of empirical beliefs. Obviously, then, in any attempt to CF justify even relatively simple beliefs, the number of jus— tificatory paths required‘to do so will simply become larger the further back towards basic beliefs we travel. The re— quirements for CF justification seem to be in conflict with one another, mutually self-defeating. It should also be pointed out that the above difficulties are only increased when one includes, as one surely ought, the justification of those observation and language beliefs which are employed in the justification of any belief. By all rights, I should justify my belief that my generous helium supplier understood by "helium" what I understand by "helium," that when I looked at the meter, the reading I seemed to see was indeed the reading of the meter, etc. In addition, the above difficulties have been amassed on the moderate 8For example, any CF properly basic belief such as, "I am being—appeared-to—redly," will only be evidence for the belief that I am seeing something red if used in conjunction with other, nonbasic beliefs such as, "When appeared-to—redly, one is justified in believing he sees something red." Since the latter is not basic, it too will require justification, and likewise for a host of other beliefs explicitly and implicitly used in justifying a simple observation belief. ‘ foundational troubles Whj Clearly support a P them all bal is not to s or that it aforementic expands whe it would nc tion would one would because th better rea The beliefs. certain In World worl our belie our autho 0f langua the Worlc‘ In fact beliefs ( Classes, justific How 154 foundationalist requirements of M—, to say nothing of the troubles which would be raised given M+ or 8*. Clearly, CF justifying all the beliefls I have which support a particular belief, since it would include tracing them all back to basic beliefs alone, is not possible. This is not to say, of course, that I have no justified beliefs, or that it would not be reasonable, having performed the aforementioned experiment, to consider my belief that helium expands when heated confirmed. It is merely to point out that it would not be CF justified, for any attempted CF justifica- tion would require ineliminable authority beliefs. After all, one would eventually come upon beliefs which one believed because their mother told them so and for which they had no better reason to believe than that she had. The situation is the same with a host of our other beliefs. Our beliefs that certain colors are called by certain names in our language, our beliefs about the way the world works, about what constitutes a good argument, and even our beliefs about our present experience are dependent upon our authority beliefs. These beliefs concern (a) the meaning of language, (b) the nature and validity of arguments, (0) how the world works, and (d) what experiences are experiences of. In fact, the complete justification of most of our empirical beliefs would require employing beliefs from all four of these classes. To see this, let us look at one more example, the justification of language beliefs. How is it that we come to understand a language? However human linguj tion, in re language fre tion and as gresses to s aries, and e learns by a teachers, 1 guage users At the of our parl is implici beliefs is tance. As of teacher these peoI authority authority secondary (teachers is certai 1y basic learned from the instruct do not r ofpa 155 human linguistic ability developed and however brains func- tion, in regard to language, it seems clear that we learn language from previous language users. This includes imita— tion and association at the earliest stages, and later pro- gresses to self-conscious methods utilizing teachers, diction- aries, and other learning tools. Throughout this process, one learns by accepting the authority of others, whether parents, teachers, lexicographers, authors or other more skilled lan- guage users. At the most basic level, we accept the linguistic usage of our parents, and could not do otherwise. This acceptance is implicit, of course, but implicit acceptance of authority beliefs is no less reliance on authority than explicit accep- tance. As we grow older, we explicitly accept the authority of teachers in these and related matters. How do we know that these people should be believed? Simply put, we accept their authority on the authority of others. Given the primary authority of parents, etc., it is rational to accept the secondary authority of those they inform us are trustworthy (teachers, e.g.). Such reliance on our primary authorities is certainly rational, but not because it is either CF proper— ly basic or derivable from such beliefs. Rather, having learned by the example, teaching, and instruction of others from the outset, we naturally continue to abide by their instructions concerning whom else we ought to learn from. We do not recount reasons why we ought to listen to the authority of P about the authority of Q; we simply accept that authority . N0 justifica authorities it is prior very small how could h beliefs by the authori source of more basic ity belief ties about not only c books, ar utilize uI At t1 proceed. we know bf tion thro Wishes to that pari by relyi Since We this pro Pervasiv \ . 9No In eithe 156 authority. P’s authority is basic,9 a given which we accept. No justification need be recited because the authority of our authorities is as basic to us as any other belief. Indeed, it is prior to most if not all our other beliefs. Unless a very small child, e.g., simply accepts what he is "taught", how could he learn at all? He has at that point no store of beliefs by which to challenge these authorities. Accepting the authoritative utterances of others has been his greatest source of verbal belief acquisitions; what, then, could be more basic to him? Moreover, we continue the chain of author- ity beliefs by accepting the authority of these new authori- ties about other authorities. This includes our acceptance not only of what is said by these authorities, but also the books, articles, and other information sources which we utilize under their direction or on our own. At this point, it should be obvious how the argument will proceed. As a matter of fact, we have learned most of what we know by accepting authorities. This is true of our educa- tion throughout life. If one decides to enter a new field and wishes to learn about it, he or she must go to authorities in that particular field. That these are authorities is known by relying on authorities who inform us that this is so. Since we have learned such a great deal of what we know by this procedure, it should be clear that authority beliefs are pervasive in our noetic structure. We not only did not, but 9Not, of course, per the last section, "properly basic" in either the CF or the Plantingian sense. also could I relying UPCr that process effect, par1 rational crl edge, and e engage in t past and/or so that we of our ac acquire mc authority one’s noet basic and paths to 1 Three mind. F: discovery was said 157 also could not have even begun the learning process without relying upon them, and we certainly could not have continued that process in any area without utilizing them. It is, in effect, part and parcel of the very method or means by which rational creatures such as we are obtain information, knowl— edge, and experience. That being the case, any learning we engage in uses knowledge based on authority procured in the past and/or the present in order to gain further knowledge, so that we not only possess the knowledge we do have because of our acceptance of authority beliefs, but also cannot acquire more knowledge without relying upon and utilizing authority beliefs. Thus it seems that ABs are essential to one’s noetic structure despite the fact they are not properly basic and cannot be supported by some path or combination of paths to properly basic beliefs alone. Three objections to this scenario immediately come to mind. First, are we not merely confusing the context of discovery with the context of justification. While all that was said above may be true, it might be maintained, you are only recounting the process by which we learn, not the process by which, once having learned sufficient information and skills to think rationally, we can then justify our beliefs. It is this post facto justification which foundationalism holds must be in accord with its understanding of justified belief for our beliefs to be truly rationally warranted. The process of learning is of psychological interest, but philoso~ phically beside the point. Perhaps endemic to epistemic tI be used in complete Pl beliefs? MI not only ut but also wi authority 1 process. utilizes a1 of them. high groun all our at of those a How v belief, f authority phrases b 1y approp response. if the hy other be order to Which, i by relyj One 158 Perhaps, but if as I have urged authority beliefs are endemic to our learning even the most basic concepts and epistemic tools, and if these concepts and tools must later be used in the justification of our beliefs, how can any complete ,pgst__;agtg justification be free from authority beliefs? Moreover, since that justificatory process requires not only utilizing knowledge learned in the past on authority but also will likely involve learning new things on authority, authority beliefs will necessarily be employed in any such process. But then, since our justificatory process itself utilizes authority beliefs, we are not, indeed cannot, be free of them. We cannot, that is, ever reach an epistemological high ground from which we can View all our beliefs, including all our authority beliefs, without relying on at least some of those authority beliefs. How would one go about justifying a particular linguistic belief, for example, without relying on things learned by authority? One can, of course, test the meaning of words and phrases by giving subjects the word or phrase in hypothetical- ly appropriate circumstances and then observing the behavioral response. By such a method, one would indeed be able to see if the hypothesized meaning was, given the accuracy of all the other beliefs utilized in the test, the real meaning. But in order to do so, one must utilize these other beliefs, most of which, if not all, are likewise beliefs which were formed only by relying upon authority beliefs. One might reply that while it is certainly true that we cannot teSt allow us to But again, 1' beliefs, we such a way ‘ paths endin those justi Only if we authority ] beliefs, a1 satisfied noted ab0\ belief ha‘ ultimate ‘ made it e creatures sort we 5 (attempts beliefs. Anoi to justi all lea: justify fore I all to CF pr from th elininai 159 cannot test all our beliefs at once, a piecemeal method would allow us to foundationally justify our beliefs one by one. But again, if we can do so only by utilizing other authority beliefs, we can never so justify them, even one at a time, in such a way that all authority beliefs have been justified by paths ending in CF foundational beliefs alone, for each of those justifying endeavors will itself use authority beliefs. Only if we can somehow justify beliefs without using any authority beliefs could we ever justify any of our nonbasic beliefs, authority beliefs or other kinds, in a manner which satisfied the strictures of CF foundationalism because, as noted above (p. 149), those strictures require that every belief have at path to foundational beliefs which are its ultimate justification. The examples above, however, have made it clear that such. could. never' be done by rational creatures who must utilize authority beliefs to learn, which sort we seem to be. All of our beliefs and the testing of (attempts to justify) our beliefs, are infected by authority beliefs. Another way of putting the matter is this. Any attempt to justify our beliefs will involve learning. Since (most) all learning' utilizes authority beliefs, all attempts to justify our beliefs will utilize authority beliefs. There- fore, authority beliefs cannot be justified solely by recourse to CF properly basic beliefs and beliefs evidentially derived from the latter alone. That is, authority beliefs are in— eliminable from our ratiOnal noetic structure and, therefore, are essentia beliefs are be amended ‘ justificatie But, i‘ to justify I which the be Must the le used in a 1: tion? And Why cannot justificat how any 0 beliefs at While our close helium he done. Tl justifice tions, at it plain] ted from and rela Will be 160 are essential to our rational beliefs. But since authority beliefs are not themselves CF properly basic, CF must either be amended or discarded. Here, it appears, the context of justification is not independent of the context of discovery. But, it might be objected, why might it not be possible to justify our beliefs without referring at all to the way in which the beliefs employed in that justification were learned? Must the learning processes by which were acquired the beliefs used in a particular justification be part of that justifica- tion? And must justification involve learning new things? Why cannot one simply lay out those beliefs which lie in a justificatory path, beliefs already possessed, leaving aside how any of them were learned and not bringing in any new beliefs at all? While this possibility might be plausible prima facie, our close examination above of a CF justificatory path for helium has revealed that it likely could never actually be done. This example made it clear that just laying out that justificatory path would require our learning new proposi- tions, and so if such learning requires authority beliefs--as it plainly seems to——then authority beliefs cannot be elimina— ted from the complete justification of even this fairly simple and relatively uncontroversial belief. A fortiori, the same will be true of more complex and questionable beliefs. A second objection can be mustered by challenging the claim that all our learning uses authority beliefs essential- ly. Does not an infant discover many things by simple exploration discover a g apparatus? or derivati seem to shc without autl 1y, at leae liefs, by v some belief but potent: is, it wor without au‘ able from that some authority fore, cou process, Procedure that it ( also for beliefs . Ind ar(lument have got the com most of ignoran 161 exploration of her environment? Similarly, do we not also discover a great number of things by virtue of our perceptual apparatus? And do not these qualify as either properly basic or derivatives therefrom? Moreover, if, as these examples seem to show, we do have some beliefs which were acquired without authority beliefs, why cannot we justify (theoretical- ly, at least) all our beliefs, including our authority be- liefs, by virtue of them alone? We would then not merely have some beliefs which did not use authority beliefs essentially, but potentially could eliminate all authority beliefs. That is, it would seem that if we can learn any beliefs at all without authority beliefs, authority beliefs are not inelimin- able from our rational noetic structure. Yet it seems clear that some of our beliefs are so learned. Consequently, authority beliefs cannot be ineliminable and are not, there- fore, counter-examples to CF foundationalism. Our learning process, in other words, is not the dire epistemological procedure outlined in the previous paragraphs, but is such that it does allow not only for properly basic beliefs, but also for beliefs derived therefrom which do not use authority beliefs essentially in that derivation. Indeed, this sort of scenario must be possible, the argument could continue, for if not, how could humankind ever have gotten to the point where reliable authority existed in the community? After all, there were no authorities during most of the evolutionary process which took mankind from ignorance to knowledge. The procedure used by evolution had previous ye tion did no ed. But 11 previous e‘ theoretica This (1) Whe actually infants authori they 1e likely Procedi 162 to be one of trial and error wherein our ancestors used their direct experience to form judgments, confirming these by future experience. Surely, authoritative teaching by the previous generation must have been employed also; each genera- tion did not have to reinvent what had previously been learn- ed. But unless all those authority beliefs were rooted in previous experience, they could not have arisen at all. So theoretically, all authority beliefs must be eliminable. This argument can be outlined as follows: (1) Humans learn some propositions by direct observation without using any authority beliefs essentially. (a) The example of infants and young children. (b) The example of adults. (Perceptual experiences, e.g.). (c) Human evolution demonstrates the utilization of such a method. (1) Humankind passed through stages where authority beliefs did not exist. (2) Knowledge was obtained during those stages. (3) Therefore, knowledge can be obtained without authority beliefs. (2) If some propositions can be learned without essential reliance on authority beliefs, all propositions can be justified by means of those propositions. (3) Therefore, authority beliefs are eliminable. Whether premise (1) holds or not is dependent upon what actually is the case in human cognition. Is it a fact that infants and young children learn some things without using authority beliefs? While it is debatable whether or not what they learn by trial and error are propositions, it is highly likely that much of their "knowledge" is gained by such a procedure. Of course, it must be combined with the language learning that propositional l employ authori prevent us fro: knowledge. If to become prop essentially on arguably to be provide us wit cient learning sible to enabl by this means significant i beliefs, that Justification epistemologist If,as I have ity beliefs, i either. What can learn witl Adult lea 163 learning that proceeds along’ with it if it is to become 10 propositional knowledge, and language learning does seem to employ authority beliefs essentially, but that does not prevent us from recognizing such pre-linguistic learning as knowledge. If, however, such learning must utilize language to become propositional and if language learning does rely essentially on authority beliefs--both claims which do seem arguably to be the case—-then the example of infants does not provide us with the case we need to demonstrate that suffi- cient learning without the use of authority beliefs is pos- sible to enable us to justify all our propositional beliefs by this means. Moreover, even if infants do learn much significant information without utilizing any authority beliefs, that in itself will not help the CF foundationalist. Justification of beliefs, at least the justification that the epistemologist is interested in, is justification for adults. If, as I have argued, they cannot learn without using author- ity beliefs, they cannot justify their beliefs without them either. What is crucial, then, is the degree to which adults can learn without using authority beliefs. Adult learning, we have argued however, is not possible without using authority beliefs essentially. Of course, many 1DLanguage learning itself employs both reliance on authority—-taking the "word" of parents about the meaning and use of words--and observation-—observing and learning by trial and error how our parents use the words we learn. It is the former area which relies on authority essentially, for one must assume that parents are using words rightly, not deceiv— ing us, etc. Of course, this "assumption" is not explicit or conscious for infants, but it is nevertheless there. of those belie: adults such con general concept to them. For t can only be P1 being equal" P1 unless one can essential reli ophical wisdom firmed withoui beliefs, it fc ted. Surely, confirmed wit significant p: 80 it seems ‘ confirm by e beliefs and, our thesis. The eVoj “10. Since and Pre-self and error, 3‘ infant and e at least as allthority, such allthorj 164 of those beliefs can be confirmed by observation, but for adults such confirmation must use the language system and the general conceptual system which their education has bequeathed to them. For this and other well known reasons, confirmation can only be partial, tentative, and with an "other things being equal" proviso. Authority beliefs cannot be eliminated unless one can directly confirm observation statements without essential reliance upon ABs, but since, as now common philos- ophical wisdom holds, observation statements cannot be con- firmed without reliance upon a broader theory or system of beliefs, it follows that authority beliefs cannot be elimina- ted. Surely, if individual observation beliefs cannot be confirmed without "tacit knowledge", an entire theory or significant part thereof cannot be confirmed without the same. So it seems that those things which adults do learn and/or confirm by observation do depend essentially on authority beliefs and, therefore, do not count as counter-examples to our thesis. The evolutionary example fares no better than the first two. Since it relies upon the idea that pre—self—conscious and pre—self—consciously rational beings can learn by trial and error, it differs little if at all from the example of infant and early childhood learning. Moreover, animals seem at least as much as humans to learn by means of parental authority, perhaps more so. Again, that the acceptance of such authority is implicit, nonconscious, and nonpropositional does not at all weaken the case that reliance on authority is essential. Nor (or the rationa process) change "beliefS") are It might I the context of has become ple context cannot whole. It doe how one came t be justified hold it. But in general, a utilizing our the latter c former. We c from our gen place by mea are essential cation. Finally learned Witt pr0p0sitiom is also hic_ things with( does hot en. that t0 be 165 essential. Nor does the "rational" nature of such a process (or the rationality of the acceptance of "beliefs" by such a process) change the fact that authority beliefs (here perhaps "beliefs") are ineliminable. It might be objected that the above reply again confuses the context of discovery with that of justification. But what has become plain, I contend, is that justification in this context cannot be separated from the learning process as a whole. It does indeed make sense to claim that regardless of how one came to accept a particular belief, if that belief can be justified by some rational procedure, it is rational to hold it. But here we are examining justification and learning in general, and since one cannot justify something without utilizing our general learning abilities, considerations of the latter cannot be divorced from considerations of the former. We cannot justify beliefs by a process which differs from our general learning process, for justification takes place by means of that process. Thus, if authority beliefs are essential to learning, they are also essential to justifi- cation. Finally, premise (2), "If some propositions can be learned without essential reliance on authority beliefs, all propositions can be justified by means of those propositions," is also highly suspect. Even if we can and do learn some things without essential reliance upon authority beliefs, that does not entail that all our beliefs can be so justified. For that to be the case, we would have to be able to (a) justify beliefs one by ‘ tical or author. most all lUStif are authority t beliefs can b: beliefs, most beliefs are no for they depe learned by mea: this argument are eliminable A third c cannot justii authority bel and science, without rely: many, perhap foundational necessary au‘ are Provisic foundational 166 beliefs one by one and (b) do so without employing any theore- tical or authority beliefs at all. But we cannot. Therefore, most all justificatory cases will utilize other beliefs which are authority belief essential. Even if perchance some of our beliefs can be justified without reliance upon authority beliefs, most cannot. Certainly higher level theoretical beliefs are not amenable to direct perceptual justification, for they depend upon layers of beliefs which can only be learned by means of authority beliefs. Consequently, it seems this argument does not manage to show that authority beliefs are eliminable. A third objection might run as follows. Granted that one cannot justify many, perhaps most, beliefs without using authority beliefs. Granted also that certain areas-~language and science, e.g.--may never be either learned or justified without relying (n1 some authority beliefs. Nevertheless, many, perhaps most, of these beliefs can be justified by foundational beliefs granting the provisional truth of the necessary authority beliefs. These authority beliefs which are provisionally accepted can then be justified by other foundational beliefs granting other provisional beliefs on authority, and so on. No entire set of beliefs will ever be justified.without recourse to some nonfoundationally justified beliefs, but whenever a belief is in doubt, it can be justi- fied by foundational beliefs in conjunction with other provis- ional authority beliefs. And this, it may be claimed, is all foundationalism needs. Such a 509 ism is viable E including every tional beliefs though not all Such a foundai foundationalis Plantinga--are have at the pr are their ult To accon slightly to believing com 1y“ establish will not, I i ingful claim liever to th. it is CF rat: does not ma} my belief f: that by tes mediately a my belief i cient groun alone suff; If their w regreSs Wh: 167 Such a scenario may be right. In fact, if foundational- ism is viable at all, it must be possible that every belief, including every authority belief, can be justified by founda- tional beliefs in conjunction with authority beliefs, even though not all authority beliefs can be so justified at once. Such a foundationalism may be the case, but that is not CF foundationalism. CF foundationalism, as we-~following Plantinga--are construing it, holds that all rational beliefs have at the present moment paths to foundational beliefs which are their ultimate guarantors. To accomodate authority beliefs, CF might be weakened slightly to the claini that someone or some group in 'the believing community, not necessarily oneself, could "rational- ly" establish beliefs in need of justification. But this move will not, I fear, rescue CF. First, if CF is to make a mean- ingful claim, it needs somehow to connect each individual be— liever to the justification of his rational beliefs. Because it is CF rational for Einstein to believe in Relativity Theory does not make it rational for me to so believe unless I get my belief from him and know he is an authority. But I know that by testimony from authorities, and so the question im- mediately arises as to how I justify (foundationalistically) my belief in their authority. If their word alone is suffi- cient grounding for Einstein’s authority, why is not his word alone sufficient authority for belief in Relativity Theory? If their word is not sufficient, we have the makings of a regress which ends in my believing someone on authority alone, for if we trace again be the 1 first trusted i beliefs do not of beliefs wou actually more pendent upon etc.) for our beliefs. Ine would also ha task which ap our argument sible for rai We stari beliefs from empirical be involving 12 also be for] are experiel rational COI authority b t0 demonstr our rationa V-3 Authoi Autho: rational ll 168 for if we trace the regress far enough into our past, it will again be the authority of parents, teachers, etc. whom we first trusted that gets us started believing. But since those beliefs do not meet CF criteria, it would seem my entire set of beliefs would be (CF) irrational. Of course, the case is actually more difficult than we have made it, for we are de- pendent upon "authoritative sources" (publishers, teachers, etc.) for our information that Einstein ever promulgated such beliefs. In order to fully CF found this belief, all of them would also have to be CF justified too. We certainly have a task which appears impossible on the practical level, and if our argument above is sound, it is also epistemically impos- sible for rational beings such as us. We started our case for the ineliminability of authority beliefs front a rational noetic structure by examining an empirical belief and then constructed a similar argument involving language procurement. Analogous arguments could also be formulated using our beliefs about what experiences are experiences of (conceptual learning) and our concept of rational cogency, but one only need show that dependence on authority beliefs is endemic to one of these areas in order to demonstrate the ineliminability of authority beliefs from our rational noetic structure. v.3 Authority Beliefs and Rationality Authority beliefs, then, remain an integral part of our rational noetic structure, so it must often be rational for person S to be the authority C ity on X (and s that undermil course, this " will utilize 51 a physicist ca position and e his work. Thi mechanics, or authority are modern ones. any sort, or activities if Clearly, we e for us withor claims. The reasonable f verifying th 169 person S to believe proposition q about subject matter X on the authority of A if S has good evidence that A is an author- ity on X (and there is no evidence to the contrary known to S that undermines A’s authority in this particular case). Of course, this "good evidence" will not be CF evidence, for it will utilize still other authority beliefs essentially. Thus, a physicist cannot verify for himself every mathematical pro- position and empirical result he uses, yet he is rational in his work. This is even more true of computer repairmen, auto mechanics, or flight controllers. In fact, arguments from authority are unavoidable for life in any society, especially modern ones. How could we vote intelligently, make plans of any sort, or carry out most any of our jobs or even family activities if we did not believe things on the word of others? Clearly, we could not. So authorities must be substantiable for us without requiring that we ourselves substantiate their claims. There must be a way, that is, to show that it is reasonable for P to believe what S says about X without his verifying the actual propositions which S asserts. V.4 CONCLUSION I have argued that authority beliefs are ineliminable from our noetic structures and that they are neither CF properly basic nor derivable from CF properly basic beliefs. Yet, since such beliefs are often rational, I agree with Plantinga that CF standards of rationality are not justifiably imposed on authority beliefs. That is, there are rational B because it is rational, howe authority on 1 ground that ae ground it in o ally grounded Our thesis is ‘, authority bel tainty or ep beliefs shoul It may highest degre ible, but s1 beliefs. Ra properly bas teaChers, e. belief in G are beliefs from PrOper use of nor 170 beliefs which are not CF rational. Plantinga has argued that this shows that such beliefs are properly basic in a nonCF way. But then we come up against the obvious objection that this justifies ad verecundiam arguments. I accept the belief B because it is asserted by an authority. Such acceptance is rational, however, only if I have good reason to accept that authority on the issue. Now although I may not be able to ground that acceptance in foundational beliefs alone, I can ground it in other beliefs, (some of which may be foundation— ally grounded), and so authority beliefs can be justified. Our thesis is just that they cannot be CF justified, for authority beliefs do not themselves supply the sort of cer- tainty or epistemic finality which proper properly basic beliefs should afford. It may be that even those beliefs that do convey the highest degree of certainty and epistemic finality are defeas— ible, but surely they are not as defeasible as authority beliefs. Rationality is consistent with authority beliefs, but that does not mean authority beliefs ought to be granted the sort of basicality that foundationalism ascribes to properly basic beliefs. But then while Plantinga’s 14 year old boy is rational in believing in God because his parents, teachers, etc. hold this belief, that does not entail that belief in God is properly basic for him. If so, then there are beliefs which are neither properly basic nor derivable from properly basic beliefs alone (i.e., without any essential use of nonproperly basic beliefs) which are nevertheless a'r sausages?» none 6 35d: .~. | - r a; .1. .. .- . - . I) 'i'J..'I.' ..:.I "LR-3.69:1: Nit-tonnes me. yd barman. I1 3}- -..--.-.-- tax-w. -. =.= vino .19vovoa . ‘-' -- -= .-.=.--"r strict no e131 '~':.: 73:13 imam basis beliefs. 4 to consider sorts of Plan-fl a coherentisii, properly basie' themselves in 171 rational to hold. It remains for us to give an account of how such beliefs can be validated within a noetic structure if they are not properly basic and are not justifiable by means of properly basic beliefs. Before doing this, however, it will be useful to consider coherentism, for it may seem that ABS and other sorts of Plantinga basic beliefs are best construed as having a coherentist justification. Indeed, perhaps Plantinga's properly basic beliefs just are coherentist beliefs disguising themselves in foundationalist attire. Plantinga basic beliefs grounded in cm in conditions thereby warra contention is justificatior need of warra cumstances (‘ of evidentiai beliefs whic sort of inf basic beliei by means of not that om circumstanc Such a beli CHAPTER VI JUSTIFYING BASIC BELIEFS Plantinga, as we have seen, does not hold that properly basic beliefs are groundless, but claims that they are grounded in circumstances. Thus, B is properly basic for S in conditions C where these conditions are such that S is thereby warranted in believing B basically. Plantinga’s contention is not that properly basic beliefs carry their own justification, are self-warranting, or simply are not in any need of warrant, but rather that in a particular set of cir— cumstances ("hard to specify exactly") they are not in need of evidential support. By evidential support, he means other beliefs which provide evidence for a belief by means of some sort of inference.1 The circumstances warranting properly basic beliefs provide justification for the latter, but not by means of inferences from the former to the latter. It is not that one deduces or induces a basic belief from a set of circumstances, but that the circumstances legitimize taking such a belief as basic and make it epistemically proper not 1The inference may be inductive, deductive, or abductive. See Alvin Plantinga, “Coherentism and the Evidentialist Objection," Rationality. Religious Belief, & Moral Commitment, ed. Robert Audi & William J. Wainwright (Ithaca, New York: Cornell University Press, 1986), p. 115. Hereafter, CEO. 172 to seek evidenc tree before me, 1 see a tree 1 such that a be belief can ex! cunstances, n6 that I am seei ing without ex I have a in a sense t] defeasible be and this evic which are CF ather falli coherentism. this very po almost indiS‘ I have 173 to seek evidence for the belief. For example, when seeing a tree before me, I am within my epistemic rights in believing I see a tree if the conditions giving rise to my belief are such that a being with my epistemic equipment who has such a belief can expect it to be true. I do not, in these cir- cumstances, need to amass evidence in support of the belief that I am seeing a tree; I am therein justified in so believ- ing without evidence. I have argued that such beliefs are not properly basic in a sense that abets the aims of foundationalism, for as defeasible beliefs, their justification can require evidence, and this evidence must consist of at least some statements which are CF properly basic. If not, one is in danger of either falling into an infinite regress or some form of coherentism. This chapter will argue that in fact it is at this very point that coherentism and foundationalism become almost indistinguishable. I have agreed with Plantinga, however, that not all beliefs can be completely grounded in CF properly basic be- liefs, and so if beliefs are to be foundationally justified, something like Plantinga’s view seems inevitable. In par— ticular, I have argued that authority beliefs, a type of Plantinga basic belief, cannot be grounded in CF basic beliefs. It is revealing, therefore, that Plantinga’s properly basic beliefs are grounded in such a way that there seems on the surface little more than a semantic distinction between them and Laurence BonJour’s coherentist observation ‘.—1 beliefS- ““96 it might well 5 Platinga baSi justified by t] Nor is thi basic beliefs which they ari observation bt tism2 and (b) system,3 as Bc to be some 5 justifying co In fact, lik produced obs: . .. th in ques cally r those b beliefs likely they Wt either conditf 174 beliefs. Indeed, as noted at the close of the last chapter, it might well seem that authority beliefs--and therewith all Plantinga basic beliefs-—are actually coherentist beliefs, justified by their place in a coherent noetic structure. Nor is this accidental, for, on the one hand, if properly basic beliefs are justified by virtue of the conditions in which they arise, as Plantinga holds, and, on the other hand, observation beliefs (a) are necessary for a viable coheren- tism2 and (b) can only be justified by other beliefs in the system,3 as BonJour holds, then one might well suspect there to be some strong similarities between the foundationalist justifying conditions and the coherentist justifying beliefs. In fact, like Plantinga, BonJour holds that spontaneously produced observation beliefs are reliable. He writes, . . . the class of cognitively spontaneously beliefs in question [observation beliefs] must be epistemi- cally reliable in relation to the subject matter of those beliefs . . . it must be very likely that the beliefs are true when they do occur or else very likely that they would occur in situations in which they would be true, where such reliability is in either case relative to some specifiable set of conditions. . . . (SEK, 122) We shall attempt to show that these similarities in the treatment of observation beliefs exist because neither foundationalism nor coherentism is an adequate epistemology. First, we shall examine just how it is, according to 2See Laurence BonJour, The Structure of Empirical Knowledge (Cambridge, Massachusetts: Harvard Universtiy Press, 1985), pp. 108, 112f. Hereafter abreviated SEK. 3Ibid., pp. 32, 84. Plantinga: nonp therefore, 00‘“ see exactly Plantinga: fr°7 beliefs. We ST tion beliefs w us in a positi beliefs and I epistemically will then perI ces between t significant 2 which issue. we concluded does not su< evidential r beliefs he (nonCF) evi coherentist afoul of P15 COherentism by themselv be created properly b: Plantinga, nonproperly basic beliefs receive warrant and why, therefore, coherentism is untenable. This will enable us to see exactly how evidential justification differs, for Plantinga, from the justification required of properly basic beliefs. We shall then elaborate BonJour’s notion of observa- tion beliefs within a coherentist epistemology. This will put us in a position to see why BonJour thinks there are no basic beliefs and Plantinga thinks there are and that they are epistemically necessary for rational belief. That comparison will then permit us to evaluate the similarities and differen- ces between these two views, to judge whether or not they are significant and, if so, which View, if either, is correct on which issue. Our ultimate aim is to show that since (a) as we concluded in Chapter III, Plantinga's view of basicality does not succeed in exempting his basic beliefs from some evidential requirements, (b) as we concluded in Chapter V, the beliefs he classifies as properly' basic actually do have (nonCF) evidential support, and (c) BonJour’s completely coherentist account of the justification of such beliefs runs afoul of Plantinga’s main objection to coherentism—-yi;., that coherentism.requires legitimating the justification of beliefs by themselves—-some alternative epistemological structure must be created which allows for the justification of Plantinga properly basic beliefs without falling into coherentism. (among other untenable epii upon three 6 "believed-on-‘ relation, and V1.11 Ih§_§t Some bel not believed one holds. well know, 0 of these ba: Plantinga, w are also pr evident prc beliefs, am properly has that a has incorrigibl more firmly 113) What 176 VI.l PLANTINGA ON EVIDENCEI BASICALITY, AND WARRANT In "Coherentism and the Evidentialist Objection to Theistic Belief,"4(hereafter CEO), Plantinga attempts to show (among other things) why coherentism is an inadequate and untenable epistemology. To demonstrate this, he elaborates upon three epistemic concepts: the basis relation (or "believed-on-the—basis" relation), the evidential supports relation, and the property of epistemic warrant. V1.11 The Structure of a Rational Noetic Structure Some beliefs, Plantinga reiterates, are basic. They are not believed on the evidential basis of other beliefs which one holds. This, he proposes, is a matter of fact. As we well know, classical foundationalism holds that only certain of these basically held beliefs are properly so held, but Plantinga, we have seen, holds that many other of our beliefs are also properly basic. Not only incorrigible or self- evident propositions, but also memory beliefs, authority beliefs, and observation beliefs are usually, if not always, properly basic. (CEO, 113-15) At the outset, Plantinga notes that a basic belief is not necessarily ". . . certain, incorrigible, unrevisable, maximally warranted, or believed more firmly than any belief that is not basic. . . ." (CEO, 113) What makes a belief basic is not some property of the “In Rationality, Religious Belief, & Moral Commitment, eds. Robert Audi and William Wainwright (Ithaca, N.Y.: Cornell University Press, 1986), pp. 109-138. belief itself: held, Lil-I n01 "spontaneously our epistemic : ments. That evidential sul tionalists wi: mately be bel Plantinga, th A secon tial support: believed on again, Plant counts as pr relation est basis of am works or ouc_ do hold some 0f beliefs, the basis 0 177 belief itself, but the way in which it is or has come to be held, yig., not on the evidential basis of other beliefs, but "spontaneously" and "naturally" in accord with the design of our epistemic systems as they operate in appropriate environ- ments. That is, :1 basic belief is not sustained by the evidential support lent it by other beliefs. While founda- tionalists will differ over what sorts of beliefs can legiti- mately be believed in a basic way, the fact is, according to Plantinga, that we do so believe many of our beliefs. A second category of beliefs is created by the eviden- tial supports relation, yiz., nonbasic beliefs or "beliefs believed on the evidential basis of other beliefs". Here again, Plantinga notes, foundationalists differ over what counts as proper evidence, but regardless, what the supports relation establishes is the belief of one proposition on the basis of another. However the evidential supports relation works or ought to work, the fact is, Plantinga emphasizes, we do hold some beliefs on the basis of others. These two sorts of beliefs, he continues, basic beliefs and beliefs held on the basis of other beliefs, are the only two sorts of beliefs foundationalism recognizes, and along with the relations holding between them, make up a noetic structure. "A person’s noetic structure is the set of propositions he believes, together with certain epistemic relations that hold among him and these propositions." (CEO, 115) To give an account of a person’s noetic structure would require not only indicating which beliefs of her’s were basic and which were "an index of dé are held with for any belief structure], a: beliefs) 511295 it.“ (CEO, 1 history“ of e acceptance . " 5 VI.12 Basis The thrl insofar as P are the has 178 and which were not, but also the following four items. First, "an index of degree of belief," (CEO, 116) for not all beliefs are held with the same degree of conviction. Second, ". for any belief B that is a member of it [one’s rational noetic structure], an account of which beliefs (and which sets of beliefs) support B, and the degree of support they provide for it." (CEO, 116) Third, there will need to be an "epistemic history" of each belief and, fourth, an "index of degree of acceptance."5 VI.12 Basis and Support The three most important aspects of a noetic structure insofar as Plantinga’s case against coherentism is concerned are the basis relation, the supports relation, and the 5In the text, Plantinga states that there are a number of other important factors in a noetic structure, but not being particularly important for his purposes, fails to mention them. In a chapter on coherentism in an as yet unpublished book on warrant, he mentions two additional elements: The "deontological history" of each belief (specifying whether the belief was formed "in accord with epistemic duty"), and an index of coherence "which measures both the coherence of the structure as a whole and for each of its members its coherence with the rest of the structure." "Coherentism". He points out that this expanded list is still incomplete. warrants 9170pe irrefleXiVer supports relat 19) Thus, sir its own Supp‘ evidence for i cannot be le( (CEO, 116-17) doubted that tion A on th clearly shou cannot be be B is based or (CEO, p. 117 will also be is evidence muon decay. SUpport one other), the SO while th 179 warrants property.6 The basis relation is asymmetrical, irreflexive, and noncircular, but not transitive. The supports relation is irreflexive but symmetrical. (CEO, 116- 19) Thus, since both are irreflexive, a statement cannot be its own support or its own basis, i.e., it provides no evidence for itself (although it logically entails itself) and cannot be legitimately believed on the basis "of itself". (CEO, 116-17) In regard to the latter, he writes, "It may be doubted that anyone is so benighted as to believe a proposi- tion A on the basis of A; but even if it could be done, it clearly should not be." (CEO, 117) Likewise, a belief A cannot be based on a belief B while at the same time belief B is based on belief A, the basis relation being asymmetrical. (CEO, p. 117) But a belief A supported by another belief B will also be support for B. As Plantinga notes, muon decay is evidence for relativity theory and relativity theory for muon decay. However, although two beliefs may mutually support one another (i.e., each can be evidence for the other), they cannot be believed on the basis of each other. So while the supports relation moves in either direction, the 6Warrant is a property of propositions which is passed along from one proposition to another by means of the latter proposition being believed on the basis of the former. Plantinga also makes mention, however, of what he calls the "gets—a1l—its-warrant—from relation." (CEO, 122) This relation is, he holds, transitive. Throughout our discussion, I will ignore this relation since, according to Plantinga, (doxastic) warrant is transferred by the "basis relation". The fact that warrant is "transitive", however, while the basis is relation is not, would seenl to indicate that a warrants relation is also in order. See fn. 8. ll asymmetry of 1 each on the b one propositi I cannot then provided by t Neither sufficient tc rule out all sitivity of ' it is possib basis of B, without beli ofstructure for, Plantir ideal reaSOI basis" relai What belief whether or proPOSitioy 180 asymmetry of the basis relation prohibits one from believing each on the basis of the other. Consequently, if I believe one proposition because of the evidence supplied by another, I cannot then believe the latter on the basis of the evidence provided by the former. Neither of these relations, Plantinga asserts, is sufficient to refute coherentism, for they do not themselves rule out all circular reasoning. This is due to the intran- sitivity of the basis relation. Given this intransitivity, it is possible for someone to (rationally) believe A on the basis of B, B on the basis of C, and C on the basis of A without believing A on the basis of A. What makes this sort of structure rationally possible, apparently, is our finitude, for, Plantinga remarks, things might be different for "an ideal reasoner." (CEO, 119) Thus, the "believed-on-the— basis" relation has to do with our actual believing, i.e., what belief B actually gave (evidential) rise to belief A whether or not we are conscious of that origin.7 7That is, it does not necessarily entail our conscious awareness of what constitutes the basis for our belief. E.g., Plantinga presents the following possibility: "Suppose I believe A on the basis of B (alone) and suppose B supports A; suppose further that I believe B does pg: support A and, indeed, believe that it supports the denial of A. A situation like this seems possible." (CEO, 116N12) Although admittedly a defective noetic situation, the fact that it is possible shows that the basis relation does not assume that one is conscious that that relation exists between two beliefs between which it does exist. If it did, the above situation would not even be a possible one, for then I would believe a proposition on the basis of another aware that I also believed the latter did not support the former. Such a psychotic belief situation seems to me not even possible. Nevertheless, will not rule finds objeCti circle will i being belith itself. Indi by the intrar V1.13 flggggy It is t makes cohere concept borr tion“ or "ac Warrant say the make a! to mak< belief 3; it is he] measur for he ces. from r S’s h< 181 Nevertheless, if this is the case, then the basis relation will not rule out the sort of circularity which Plantinga finds objectionable in coherentism, for each belief in the circle will have evidential support without any one belief being believed, directly or indirectly, on the basis of itself. Indirectly believing B on its own basis is blocked by the intransitivity of the basis relation. V1.13 Warrant It is the transitivity of warrant which, for Plantinga, makes coherentism untenable. He states that by "warrant" (a concept borrowed from Roderick Firth), he means "justifica- tion" or "acceptability". He continues, Warrant is a normative or evaluative property; to say that a belief has warrant for a person S is to make an appraisal, an evaluative judgment. It is to make a positive appraisal; it is to say that the belief in question has positive epistemic status for S; it is to say that the way in which this belief is held and sustained, in S’s noetic structure, measures up to the appropriate norms or standards for holding that sort of belief in S’s circumstan— ces. It is to say that S’s noetic structure suffers from no defect or blemish or malfunction by way of S’s holding this belief in the way he does. More particularly, it is to say there is no defect or blemish in S’s noetic structure by virtue of the way this belief is related to S's other beliefs and to 5’5 experience. (CEO, 119) Warrant, Plantinga holds, comes in degrees and the degree of warrant a nonbasic belief has depends on the degree of war- rant possessed by the warranting beliefs and the strength of the supports relation between warrantors and warrantee. (CEO, 120) Importantly, warrant is transitive but does not increase when warrant i to another. ( the basis reli A is warrante will be warra warrant is i! mean A (and t at all. (CE( To see i Plantinga’s , untangle tht supports rel what condit warranted. argument th; We shall th coherentism V1.131 Bag Is it Prevents tr mutually s keeps them That is, j thinking, the Suppm 182 when warrant is transferred from one belief or set of beliefs to another. (CEO, 122) Because warrant is transitive, unlike the basis relation, if A is warranted by B and B by C, then A is warranted by C. Thus, if C is warranted by A, then A will be warranted by A. But since, like the basis relation, warrant is irreflexive and asymmetrical, such a circle will mean A (and every other member of the chain) has no warrant at all. (CEO, 122) To see this more clearly, it will be helpful to develop Plantinga’s argument in more detail. First, we shall try to untangle the relations which hold between the basis and supports relations and warrant in order to make clear exactly what conditions must be met for a nonbasic belief to be warranted. Second, we shall need to examine Plantinga’s argument that warrant is not increased by warrant transfer. We shall then be able to see clearly why Plantinga rejects coherentism. VI.131 Basis, Support, and Warrant Is it the asymmetry of the warrants property which prevents the supports relation from providing evidence for two mutually supporting beliefs at the same time, i.e., which keeps them from being believed on the basis of each other? That is, is the asymmetry of the basis relation a result of the asymmetry of warrant? Probably not to Plantinga’s way of thinking, for his explanation of the irreflexiveness of both the supports and basis relations makes no mention of warrant, and the basis tive“ one 11 transfers (or not) in a par on the basis did not beli supports A. A has no war] I believe or I actually . others. Si seems to be B whether 0 and B A, I basis of A to do so, reflexive. while B we warrants p Objective relations these latt Warre believed warrant v Warranted rightly r 183 and the basis relation is a doxastic category, not an "objec— tive" one like the warrants property. That is, warrant transfers (or not) regardless of my belief that it does so (or not) in a particular case. If B has warrant and I believe A on the basis of B, then A has warrant for me. But even if I did not believe A on B, A would still have warrant, for B supports A. Moreover, if B does not actually support A, then A has no warrant even if I believe it does. But which beliefs I believe on the basis of others clearly depends upon which I actually do believe—-conscious1y or not-~on the basis of others. Similarly, the supports relation, for Plantinga, seems to be an "objective" relation, proposition A supporting B whether or not I believe it to. So although A supports B and B A, I cannot believe A on the basis of B and B on the basis of A simply because it is irrational, pure and simple, to do so, i.e., because the basis relation itself is ir— reflexive. That A cannot warrant B (in a noetic structure) while B warrants A is due to the irreflexiveness of the warrants property, pure and simple. So, then, these are objective properties (asymmetry, etc.) of properties and relations (warrants, basis, and supports, respectively) which these latter have by virtue of their "natures". Warrant, then, will not accrue to a belief because it is believed on the basis of a belief which supports it, but warrant will be transferred to a belief because another warranted belief supports it. The former, therefore, will rightly be believed on the basis of the latter. So I can believe A on ' belief of A w A, although B B has no warr by the suppor But warr on the basis believing A warrants A warranted if which suppo: warranted by or not I b however, th “Suppose I supports A; A and, ind (CEO, ll6N] is possible of rationa "rational“ \ aHere relation Supports 1 must be s which dis; It not be the Warral One Could of Warren well as a 184 believe A on the basis of B, but only if B warrants A is my belief of A warranted. And B only warrants A if B supports A, although B may support A without warranting it (if, e.g., B has no warrant). So warrant transfer is sometimes carried by the supports relation, but not necessarily so.8 But warrant transfer is also, in another sense, dependent on the basis relation. As stated above, I am warranted in believing A on the basis of B only if B warrants A and B warrants A. only if B supports A. Thus, my' belief A is warranted if believed on the basis of a warranted belief B which supports A. In an "objective" sense, belief A is warranted by belief B if B is warranted and supports A whether or not I believe A on B. But Plantinga seems to think, however, that the following sort of situation is possible: “Suppose I believe A on the basis of B (alone) and suppose B supports A; suppose further that I believe B does pg: support A and, indeed, believe that it supports the denial of A." (CEO, 116N12) Although such a doxastic situation, he feels, is possible, it is a defective one and, therefore, falls short of rationality ("nondefectiveness"). So a belief, then, is "rational" only if believed on the basis of a belief which 5Here it seems as though what is needed is a warranting relation in addition to the supports relation, for if the supports relation does not necessarily entail warrant, there must be some property of those support relations which do which distinguishes them from those which do not. So would it not be helpful to distinguish these by calling the former the warrant relation? Actually, two possibilities seem open. One could hold that some supports relations have the property of warrant transfer or that there is a warranting relation as well as a supports relation. supports it ( the "objectiV in addition t not transfer] Plantint If I l strongl basis 0 source I belie believe support warrant warrani offend: In a footnoi "propositio warrant“.‘o warrant, b1 (rationalit warrant tr: warrant tr: not just i Supports A 0f" relati entail th relation I Warrant , 185 supports it (and, therefore, can warrant it), but more than the "objective" relation of support must be present; I must in addition believe that B supports A. Otherwise, warrant is not transferred for me. Plantinga also presents the following example. If I believe propositions A, B, and C, which strongly support D, but do not believe D on the basis of those propositions, then if D has no other source of warrant, it has no warrant for me; and if I believe D on the basis of those propositions but believe that they do not support D or that they do support the denial of D, then, once more, D has no warrant for me-—or, more conservatively, it has less warrant than it would have in the absence of the offending beliefs.9 In a footnote he adds that such beliefs as D may have Firth’s "propositional warrant," but not what he calls "doxastic warrant".10 Propositional warrant, then, "undergirds" doxastic warrant, but does not imply it. So while the legitimacy (rationality) of the basis relation depends on the reality of warrant transfer (both propositional and doxastic), doxastic warrant transfer also depends on the actuality of belief, and not just belief of A on the basis of B, but belief that B supports A. Thus, the basis relation ("belief—on—the—basis— of" relation) does not entail warrant and warrant does not entail the basis relation; warrant entails the supports relation, although the supports relation does not entail warrant. 9From a chapter entitled, "Coherentism", in a forthcoming book on warrant. 10Ibid. is warranted odd situatic path--for ti on the one nalist view apparently nonbasic be V1.132 Wal As st; increase b a given pre 0f other g cannot exC Plantinga prOPOsitie 186 In summary, it seems that for Plantinga a nonbasic belief A is doxastically warranted (and, therefore, rational) iff (a) A is believed on the basis of a belief B which actually supports it, (b) it is believed that B actually supports A, (c) B is actually warranted (not just believed to be war— ranted), and (d) B is not warranted by A or any belief which is warranted by A. A can, however, support B and, in a rather odd situation, B can have A in its "believed-on-the-basis-of" path—-for the basis relation is not transitive. Thus, while on the one hand Plantinga seems to take a completely exter- nalist view towards basic beliefs, on the other hand. he apparently accepts at least some internalist requirements for nonbasic beliefs. VI.132 Warrant Transfer As stated above, Plantinga holds that warrant does not increase by means of warrant transfer. "If my warrant, for a given proposition, arises from my believing it on the basis of other propositions, then my warrant for that proposition cannot exceed my warrant for any of those other." (CEO, 120) Plantinga quickly adds that he is not claiming that if a proposition’s warrant derives from "believing it on the basis of B“ . . . , Bn, then my warrant for A cannot exceed my warrant for the conjunction of the Bi." (CEO, 120) While a natural and likely possibility, he adds, the foundationalist is not bound to it. In addition, a proposition may possess higher warrant than some of its supporting beliefs, but only if so warrant The four if A is the war] 317 he h derived then it (Perhap derived cannot possess 121) "Warrant tra Plantir seems to in) which are 1 unites them statement 5 at a New Ye; that Eleanc edge that clearly, t Paul was a1 011 the ba: Plantinga may then Possesses Sentence f011ows, quotation Chain! fC 187 if so warranted by other of its supporting beliefs. The foundationalist need not hold, therefore, that if A is believed on the basis of B“ . . , B“ then the warrant of A cannot exceed that of any of the B1; he holds instead that, if A’s warrant for S is derived entirely from its being based on the B“ then its warrant cannot exceed any of theirs. (Perhaps he will add that, if A’s warrant is partly derived from its being based upon the B1, then A cannot receive more warrant from them than is possessed by the least warranted of the By) (CEO, 121) "Warrant transfer“, in other words, does not increase warrant. Plantinga’s position here is confusing. The first part seems to imply that A is being warranted by a set of beliefs which are not necessarily connected to each other. What unites them is their support of A. An example preceding this statement seems to bear this out. A belief that one saw Paul at a New Year’s Eve party is supported by memory and testimony that Eleanor was at the party in conjunction with the knowl— edge that she seldom attends parties without Paul. Here, clearly, these two sources of warrant give the belief that Paul was at the party more warrant than that belief would have on the basis of one of these sources alone. Moreover, as Plantinga then concludes, belief that Paul was at the party may then have more warrant than either supporting belief possesses itself. (CEO, 120—21) The first half of the first sentence (up to the semicolon) in the above quote then follows. But the contention in the second half of the quotation would only be the case if the evidence were a linear chain, for surely if A is warranted by B“ . . ., B,and the B, are not connected linearly, then A has not only as much the former. substantiati‘ bolster the belief. SC warranting s B, is linear B, provides warrant as ' The me own sources warranted s In fact, th prior belie the set i connection that Elear attends pa to the be] 188 warrant as the most warranted B,, but also the additional warrant which the other B, lend it by their confirmation of the former. If that were not the case, less warranting but substantiating evidence would never be presented in order to bolster the evidence provided by the most highly warranted belief. So Plantinga. must. be thinking of each. of these warranting sets as independent chains of warrant in which each B,is linearly connected with the others. So if one chain of B, provides all the warrant for A, then A has only as much warrant as the least member of the chain. The members of the chain would each perhaps have their own sources of warrant; i.e., they would not necessarily be warranted solely by the logically prior belief in the chain. In fact, they may get none of their warrant from the logically prior belief in the chain. But since the warrant provided by the set is provided. by 'virtue of the logical or factual connections existing between the members (e.g., Bétestimony that Eleanor was at the party; Ig=knowledge that she seldom attends parties without Paul), then the warrant transferred to the belief in question can only be as strong as that of the weakest of the B,. "Linear" here need only be construed logically, and indicates that each warranting set of beliefs is to be considered independent of the others. But of course, that is only one of many possibilities. Why would not the cumulative warrant of two or three sets of warranting beliefs give the warrantee more warrant than any of the individual sets or their individual members have? Plantiné warrant. SUI any of its in warrant, he 1 state of aff of warrant e .. the deg] basis of whi relation ho warrant is beliefs and latter can the former supports re once it he must in th, warrant. SUpports Progeny w] level non} it in a fe Case, the leVel nor the Supp establis1 a first as the m, 189 Plantinga might reply that we are confusing support with warrant. Support may well be stronger for a proposition than any of its individual supporting beliefs or belief chains, but warrant, he might hold, cannot. But this would seem a strange state of affairs, for as Plantinga himself notes, "The degpee of warrant enjoyed by a nonbasic belief . . . [depends on] . . . the degree of warrant enjoyed by the propositions on the basis of which it is believed and the strength of the supports relation holding between it and them." (CEO, 120) Since warrant is a function of both the warrant of supporting beliefs and the supports relation, one would think that if the latter can increase by means of horizontal additions, so can the former. Moreover, even if one were to hold that the supports relation can only maintain warrant, not increase it once it has been established, still, the supports relation must in the last analysis, be responsible for the strength of warrant. For consider, it must be the strength of the supports relation between basic beliefs and their direct progeny which establishes the warrant enjoyed by the lowest level nonbasic beliefs, for from where else would they obtain it in a foundationalist noetic structure? But if this is the case, then surely the strength of the warrant for these first level nonbasic beliefs must be a function of the strength of the supports relation whereby it is both transmitted and established at this level. One might hold, of course, that a first level nonbasic belief can have only as much warrant as the most warranted basic belief which supports it, but why could not the dently SUPPO cases, than Surely, warre other things support the) increased b) support whit Perhap Plantinga t above (p. 1 My wa entire and E their: gener sitio on t1 warra Bus suppo than It may we] disjunctie the case, 190 could not the warrant for a nonbasic belief which was indepen— dently supported by two basic beliefs be stronger, in some cases, than that of either of the basic beliefs alone? Surely, warrant for nonbasic beliefs must be a function, among other things, of the diversity and breadth of the evidential support they have. So one would think that warrant can be increased by virtue of increasing the diversity of evidential support which a belief enjoys. Perhaps it is considerations of this sort. which led Plantinga to remark that the first sentence in the quotation above (p. 187) is "clearly an error". He continues, My warrant for a disjunction B1 v B2 may derive entirely from its being believed on the basis of B1 and B“ but its warrant may nonetheless exceed theirs. About all the foundationalist can say generally here is that [if] the warrant of a propo- sition A derives entirely from its being believed on the basis of the 11’s [warrant?], then A’s warrant will be a function of the warrant of the Bys [sic] together with the degree to which they support A; and in no case can A have more warrant than that of the disjunction of thelx’s [warrant?]11 It may well be that no belief can have more warrant than the disjunction of its warranted supporters, but is it not also the case, given our discussion above, that a belief can have more warrant than its most highly warranted warrantee, i.e., that its warrant can be increased by the warrant granted to it by less warranted warrantees, and that this "total warrant" can amount to more than the warrant of even the highest warranted warrantee? So, while "vertical" warrant cannot be greater warri themselves. any of the p. would be to Despite Plantinga’s increase ju Emphasis mi warrant tre from divers V1.133 WaJ In or view of wa1 repeat pa: Plantinga' Plant nonbasic 1 Structure basis of means of 191 increased, is it not possible that warrant can be increased by virtue of additional "horizontal" warrantors? Thus, the cumulative evidence that Jones murdered Smith might have greater warrant than any of the individual pieces of evidence themselves. It might, that is, be more rational to give up any of the pieces of evidence considered individually than it would be to give up the conclusion. Despite these difficulties, one can still agree with Plantinga's main contention, namely, that "Warrant does not increase just by virtue of warrant transfer." (CEO, 121. Emphasis mine.) If warrant can be increased, it is not by warrant transfer alone, but by the accumulation of warrant from diverse sources. VI.133 Warrant & Basicality In order to present a complete account of Plantinga's View of warrant in a rational noetic structure, let us briefly repeat part of our discussion in chapter III, utilizing Plantinga’s own development of these same points in CEO. Plantinga distinguishes properly basic beliefs from nonbasic beliefs by noting that in a foundationalist noetic structure they will (a) be basic (i.e., not believed on the basis of other beliefs) and (b) will receive no warrant by means of being believed on the basis of other beliefs (i.e., ". . . proper basicality, in contrast to the basis relation, is a source of warrant."). (CEO, 122) Nonbasic beliefs receive warrant through the "believed—on—the—basis—of" and the support rela‘ warrant, whai Plantinga's beliefs by (CEO, 123) basically, 1 structure. function of chapter III C such tha‘ believed wi V1.14 M Cohere Plantinga increased case, cohe defective. which aris 192 support relations. Since basic beliefs do not so receive warrant, what is the means by which warrant accrues to them? Plantinga's answer is that. warrant is conferred on basic beliefs by the circumstances in. which they are believed. (CEO, 123) If the circumstances are proper for believing B basically, then B is a properly basic belief in my noetic structure. On the level of basicality, then, warrant is a function of the circumstances of the believing. As we saw in chapter III, there will be some R which specifies conditions C such that in those conditions belief B can properly be believed without evidence, i.e., basically. V1.14 Plantinga’s Rejection of Coherentism Coherentist accounts of warrant would be cogent, Plantinga argues, if warrant were reflexive or could be increased by warrant transfer alone. Since neither is the case, coherentist accounts of warrant, he holds, are clearly defective. To show the illegitimacy in coherentist accounts which arises from the irreflexiveness of warrant, he considers a case in which a belief Aois believed "solely" on the basis of A“ A,solely on the basis of A“ and so on to A“ and Anis then believed solely on the basis of AM If a belief is "directly warranted" by another, Plantinga continues, it is both believed on the basis of the latter and receives all its warrant "by virtue of being believed on the basis of" the latter. (CEO, 121) Since no belief can get its warrant from itself (irreflexivity) and since warrant is transitive, it is clear that S‘ at all. (CE The sec is that it warrant, thi As we have 1 possible, a regardless does not er remains fat Given considered, tion. He 5 as transfe relation 1 He is ratio circu 193 clear that such a "circular chain" of beliefs has no warrant at all. (CEO, 121-22) The second difficulty with coherentism, Plantinga argues, is that it seems to hold that warrant transfer increases warrant, that the basis relation somehow generates warrant. As we have seen, however, he holds that this is clearly not possible, and so the circularity of warrant in coherentism, regardless of how large or complex that circularity might be, does not enhance warrant at all. Consequently, coherentism remains fatally flawed by the first difficulty. (CEO, 124) Given the untenability of coherentist justification so considered, Plantinga explores an alternative characteriza- tion. He suggests that the coherentist does not see warrant as transferred by the basis relation, but that the basis relation is a "source of warrant". He is not holding that the basis relation in a rational neotic [sic] structure can sometimes be circular; he is instead pointing to a condition under which a belief is properly basic--a condition under which it is rational to hold a belief without accepting it on the evidential basis of other beliefs. (CEO, 125) This amounts, Plantinga asserts, to a condition of proper basicality. ". . . A belief B is properly basic for a person S if and only if B coheres with S’s noetic structure——or with the rest of the [sic] S’s noetic structure, if S already holds B." (CEO, 125) So for a "pure coherentist", then, all beliefs in a noetic structure would be basic, getting warrant not from other beliefs, but by virtue of their coherence with the rest of the structure. He notes that a coherentist might attenuate he some beliefs on the basi: really char: 12W 126) He perspective colors as a properly be coherence 1' coherentisi This foundation dationalis basic, the coherence, be intros accompani, foundatio 194 attenuate her view and allow "local foundationalism" whereby some beliefs would receive warrant by virtue of being believed on the basis of other beliefs, but, he concludes, "What is really characteristic of coherentism is not a View about the transmission of warrant but a view about its source." (CEO, 126) He then notes ironically, "Seen from the present perspective, therefore, the coherentist reveals her true colors as a foundationalist with unusual views about what is properly basic." (CEO, 126) Moreover, for the coherentist, coherence is "the gply source of warrant; this is the central coherentist claim." (CEO, 126) This contrasts sharply, Plantinga concludes, with the foundationalist’s View of warrant source. Although foun- dationalists differ on what sorts of beliefs are properly basic, they agree in not finding the source of warrant in coherence, but in experiences of certain sorts, whether they be introspective, perceptive, memorative, or that which accompanies self-evident beliefs. (CEO, 126) For the foundationalist, then, it is the origin of a basic belief within an experiential context, which in turn satisfies the appropriate conditions for such an individual’s believing that sort of belief, which provides warrant for the belief. This, Plantinga feels, is a far more plausible view than coheren- tism, however the claims the latter makes are construed. VI.I5 My Let us now summarize Plantinga’s view of warrant as it relates to properly basic beliefs in order to provide a concise characterization for comparison with BonJour’s schema for a coherentist justification of observation beliefs. (1) Properly basic beliefs do not arise by in- ference fronlother'beliefs, but "spontaneously" in accord with our epistemic nature. (2) Properly basic beliefs are, nevertheless, warranted by virtue of the circumstances in which they arise. If such circumstances are in accord with some rule R which warrantedly holds that beliefs of the kind in question are basic in conditions of the existent kind, then one is warranted in so believing basically. There exist then, (a) Basic beliefs of some class K. (b) Warranting conditions C. (c) A rule or law which states that for a certain class of believers in conditions C beliefs of class K are basic. VI.2 EQflQQHB1AH_QQflEBEflIl§I_QB§EBEAIlQH_BELlEE§ Plantinga has argued that coherentism cannot justify beliefs, that without basic beliefs, warrant will not accrue to any nonbasic beliefs. Laurence BonJour holds, contrari— wise, that since there are no basic beliefs, coherentist arguments must be able to justify belief if we are to avoid sceptism. He recognizes, however, that coherentist systems must somehow be influenced by the world, that a system must include within its justifying arguments some input which con— nects it to the world. The means by which this occurs, he holds, is "observation beliefs." As he analyzes them, however, they properly basic that is just BonJour, of 01 properly basic Observation be construal of o cation of then pares with Pla an account is v1.21 m For many might ring tr labels “cohere between the a transfer that and coherem-_jL View concerns a COherentist of 0bservati, thereby Provj amJues, cohe: Objection", I inDUt from t1 be otioSe; a1. tion of the v 196 however, they take on a striking resemblance to Plantinga’s properly basic beliefs; indeed, Plantinga holds that in fact that is just what they (or at least some of them) are. BonJour, of course, would likely counter that Plantinga’s properly basic beliefs are better seen as central coherentist observation beliefs. This section will analyze BonJour's construal of observation beliefs and his coherentist justifi— cation of them in order to see (a) how such an account com- pares with Plantinga’s and (b) whether or to what extent such an account is tenable. VI.21 Observation Beliefs and Coherentism For many "classical" coherentists, Plantinga’s analysis might ring true, but BonJour has developed a view which he labels "coherentist" and which seems at the least a compromise between the alternative views on warrant source and warrant transfer that Plantinga has suggested exist in foundationalism and coherentism respectively. The relevant part of BonJour’s View concerns the role he believes observation must play in a coherentist system. Unless coherentism can give an account of observation beliefs which hooks them to the world and thereby provides for input from the world to the system, he argues, coherentism will not be able to answer the "input objection", yiz., that unless a coherentist system has some input from the world, its claim to be true of the world will be otiose; and even if it is a true representation or descrip- tion of the world, that will be a matter of pure luck, and so epistemically ‘ task is to sho way of observa In develo coherentist f belief is arri the foundatior at inferential makes them, if dent upon othe exclude them i justification BonJour makes of discovery' But wha' observation count toward: corresponden time being d justificatic tist justifi noninferent; tiSt observ 197 epistemically unjustified anyway. (SEK, 108) Thus, his main task is to show how a coherentist system obtains warrant by way of observation beliefs. In developing his View of observation beliefs within a coherentist framework, BonJour first distinguishes how a belief is arrived at from its justification. He agrees with the foundationalist that observation beliefs are not arrived at inferentially, discursively, or by any other means which makes them, insofar as we are considering their origin, depen- dent upon other beliefs. However, he continues, that need not exclude them from a need to rely upon other beliefs when their justification is the point at issue. (SEK, 112-113) Thus, BonJour makes the traditional distinction between the "context of discovery" and the "context of justification." But what sort of a belief would count as a coherentist observation belief? That is, what sort of a belief would count towards the truth of a coherent system, i.e. towards its correspondence with the external world, while at the same time being dependent upon beliefs within that system for its justification? It would have to be a belief "whose coheren- tist justification depended in some way on the manner of its noninferential origin. . . . And this is exactly what coheren- tist observation ought to look like." (SEK, 113) Taking a cue from Sellers in, "Empiricism and the Philos- ophy of Mind," BonJour develops a coherentist view of observa- tion beliefs in which they act as a check on the coherentist system, but are themselves justified by beliefs which are a part of that SY way in which 121' in their lUStj form of an arg is necessary t ing the belie: the premises e To bring uses the exam such a belief certain, reas visual belief a 'medium-si: there must b takes place t ical observa law of natur Observer’s c Vers general (SEK, 118) The St: be as follov part of that system. This is possible, he argues, becausc the way in which these beliefs are arrived at plays a crucial role in their justification. This justification will be in the form of an argument, naturally, and to form that argument it is necessary to specify certain "crucial facts . . . concern— ing the belief and its context which can plausibly serve as the premises of a justificatory argument." (SEK, 117) To bring to light the sort of "facts" he has in mind, he uses the example of seeing a red book on the desk. First, such a belief must be "a cognitively spontaneous belief of a certain, reasonably definite kind K,, . . . that . . . is a visual belief about the color and general classification of a ’medium—sized physical object.’" (SEK, 117-18)12 Second, there must be a set of conditions in which the observation takes place that are the right sort of conditions for a verid— ical observation of this kind. Third, there must be "a true law of nature" about the observer (or, more accurately, the observer’s class of observers) to the effect that such obser— vers generally have veridical observations in such conditions. (SEK, 118) The structure of the justificatory argument, then, will be as follows: (1) I.have a cognitively spontaneous belief of kind IQ that there is a red book on the desk. (2) Conditions C; obtain. (3) Cognitively spontaneous visual beliefs of kind ”Bonjour points out that "Visual" is meant to indicate not the cause, but the "intrinsic character and content of the belief." (SEK, 118) So visual hallucinatory or dream beliefs are included. K,in Therefore is very 1 Therefore (SEK, 118 Or, more gener (1) I he P wt (2) Cone (3) CogI cone Therefore Therefor BonJour premises, it course to any the crucial d. is that the j he sees a re foundational and context Moreover, w} necessary is more precise that is char 0f juStifie 1y, they w 199 K,in conditions C,are very likely to be true. Therefore, my belief that there is a red book on the desk is very likely to be true. Therefore, (probably) there is a red book on the desk. (SEK, 118) Or, more generally, (1) I have a cognitively spontaneous belief that P which is of kind K. (2) Conditions C obtain. (3) Cognitively spontaneous beliefs of kind K in conditions C are very likely to be true. Therefore, my belief that P is very likely to be true. Therefore, (probably) P. (SEK, 123) BonJour notes that in addition to justifying these premises, it will be necessary to justify them without re- course to any basic beliefs. (SEK, 118) As BonJour sees it, the crucial difference between coherentism and foundationalism is that the justification of the "original belief" (here, that he sees a red book) is not "intrinsic or primitive," as in foundationalism, "but is rather dependent on the background and context provided by my other beliefs." (SEK, 119) Moreover, what shows that such a justificatory argument is necessary is that it is just the lack of such an argument (or, more precisely, part thereof, i.e., one of the above premises) that is characteristic of observation beliefs that fall short of justified belief, i.e., knowledge. (SEK, 119) So for BonJour, although observation beliefs will arise spontaneous- ly, they' will require justification, just like all other beliefs, and this justification will take the form of the above outlined argument. Moreover, each of the above premises will be justified without recourse to "basic beliefs", but rather by virtue of their coherence with the rest of one’s beliefS- From our 1 between BonJOL' beliefs and 1 observation be a belief B is a rule R whic quires that tr belief P with that premise 1 to his rule beliefs as th in conditions 0n the 1 seem to be (e acquired in : BonJour want be placed wi- and (b) that have "cognit General and Plantinga dc ity. For P: a Person wh beliefs. From our perspective, there is a remarkable resemblance between BonJour’s coherentist justification of observation beliefs and Plantinga’s justification for taking (some) observation beliefs as properly basic. Plantinga allows that a belief B is basic if believed in conditions C and there is a rule R which specifies that B is basic in C; BonJour re— quires that there be a justificatory argument for observation belief P with premises such as (1)—(3) above. It is clear that premise (2) corresponds to Plantinga's conditions C, (3) to his rule R,13 and (1) to his characterization of basic beliefs as those which arise "by nature", without inference, in conditions such as C. On the surface, the only real differences between them seem to be (a) that Plantinga wants to call a belief that is acquired in such an epistemic situation properly basic while BonJour wants to insist that the justificatory circumstances be placed within a coherentist argument for each such belief, and (b) that BonJour wants to require of the observer that he have "cognitive access" to the needed justification, both in general and for each particular case (SEK, 123), whereas Plantinga does not make that a requirement for proper basical- ity. For Plantinga, proper basicality does not require that a person who believes he is seeing a red book in conditions 13Plantinga himself only talks of conditions which warrant basic beliefs, but clearly if certain conditions warrant certain beliefs, the fact that they do is a "rule" or "law" to that effect. c either know which justifies conditions C e belief require insist, on the requires is nc if it existed, the observati involve circu that without : unjustified. Underlyi experiences c only beliefs ist must hol ience or cc required bel basic. Cons a luStifica1 tification e be warrante, and again w. 201 C either know that there is some "rule" of "law of nature" which justifies his so believing or even that he be aware that conditions C exist, but for BonJour a justified observation belief requires knowledge of both. Moreover, Plantinga would insist, on the one hand, that such a justification as BonJour requires is not necessary, and, on the other hand, that even if it existed, it would not succeed in providing warrant for the observation belief because in the long run it. would involve circular reasoning. BonJour would respond, however, that without such a justification, the belief would simply be unjustified. Underlying these difference is Plantinga’s belief that experiences do justify beliefs and BonJour’s conviction that 4 only beliefs can justify beliefs.1 of course, a foundational- ist must hold that some beliefs can be warranted by exper- ience or conditions, namely, basic beliefs, for if they required beliefs for their warrant, they simply would not be basic. Consequently, it is clear that for a foundationalist a justificatory argument will not be necessary for the jus- tification of a basic belief, for then the basic belief would be warranted by other beliefs (the premises of the argument) and again would itself not be basic. But if no justificatory 1“This assumption seems at least implicitly involved in his antifoundationalist argument (SEK, 30—33). Cf. also SEK, 84, "More specifically, there is no way for the foundational— ist's allegedly basic empirical beliefs to be genuinely justified for the believer in question without that justifica- tion itself depending on further empirical beliefs [emphasis mine] which are themselves in need of justification.“ it to be basic conditions can be justified b have an infini Of coherentist Given tht central impor' beliefs such z entism presen involved in 4 circular? In of premises ( If both are p temologies b one or the 0' one alternat The questiOI Grounding b. hardly be 5 argue, howe) of Premises Simply do n native to b Oped in 01 Properly b, 202 argument is necessary for a belief to be properly basic, then clearly the believer need not be aware of such an argument for it to be basic for him. On the other hand, if experience or conditions cannot themselves ground beliefs and beliefs must be justified by other beliefs, then obviously we must either have an infinite regress of justifying beliefs or some form of coherentist justification must be valid. Given these fundamental differences, two issues are of central importance: First, can beliefs be grounded in non- beliefs such as experiences or conditions; second, can coher~ entism present a cogent argument that the sort of circularity involved in any coherentist justification is not viciously circular? In particular, can BonJour provide a justification of premises (1)-(3) above which is both coherentist and valid? If both are possible, then we seem to have two different epis— temologies both of which are, so far at least, tenable. If one or the other is not possible, we have at least eliminated one alternative and shown that the other is, so far, tenable. The question of whether or not experiences are capable of grounding beliefs has been discussed extensively and could hardly Ix; satisfactorily treated ix] this study. I shall argue, however, that BonJour’s arguments for the justification of premises (1)-(3) either entail foundationalist elements or simply do not work. Consequently, I will conclude, an alter- native to both coherentism and foundationalism must be devel— oped in order to give an adequate account of "Plantinga properly basic beliefs", which type of belief, we have seen, includes BOHJO‘ v1.22 m; A crucial the developmen fication of th observation be he must show t es is potenti‘ that a recour He begin tion of some generally re observers. holds, need justificatio to basic bel: tion must co: (SEK, 125) tive inducti appeal can Previous pa % obvious rez arrived at the justif; question . 203 includes BonJourian "observation beliefs". VI.22 BonJour’s Justification of Observation Beliefs A crucial element in BonJour’s defense of coherentism is the development of a general approach for a coherentist justi— fication of the premises in his argument schema for justifying observation beliefs. Such a development is necessary because he must show that a coherentist justification of these premis- es is potentially available for every observation belief and that a recourse to foundational beliefs is never necessary. He begins with premise (3) (SEK, 124ff), the justifica- tion of some law which states that beliefs of kind K are generally reliable in conditions C for a certain class of observers. A coherentist justification of such laws, he holds, need not be significantly different from any other justification, except, of course, that appeal cannot be made to basic beliefs. (SEK, 124—25) Consequently, all justifica— tion must come from "within the observer’s system of beliefs." (SEK, 125) As he goes on to state, if, for example, enumera- tive induction is the means used to justify a particular law, appeal can be made ". . . only to the truth or falsity of previous particular beliefs of the kind in question as as- sessed from within the system." (SEK, 125) Although for obvious reasons such beliefs cannot "in general" have been arrived at inductively, this is irrelevant, for again, it is the justification, not the origin, of the beliefs that is in question. So the "elements" of the justifying system will need to have empirical cogn of course attempt to jus to, and these V it is also cl< once, or some' to justify tr longer be any also seems cl a belief only thereby of ne upon the ori justifying b that this am unwarranted {B} by {C}, eventually 5 previously \ since that Plantinga a Set of belie be in PIS need to have been in place before a "genuinely rational empirical cognition" is possible. (SEK, 125) Of course, given BonJour’s View, it is clear that in any attempt to justify a belief, other beliefs will be appealed to, and these will of necessity be "prior" beliefs. Moreover, it is also clear that one cannot justify all her beliefs at once, or somehow go outside her beliefs altogether in order to justify them. Once one did the latter, there would no longer be any beliefs to appeal to for justification. But it also seems clear that if one is going to attempt to justify a belief only by recourse to other previously held beliefs, thereby of necessity (barring an infinite regress) coming back upon the original justifiers for the justification of other justifying beliefs, one runs smack into Plantinga’s argument that this amounts to circular reasoning of an undesireable and unwarranted sort. If, for example, P is warranted by {B} and {B} by {C}, and so on, I can only stop an infinite regress if eventually some set of warranting beliefs is warranted by some previously warranted set in P's warranting path. But then, since that set is also warranted by that same path, as Plantinga argues, it is warranted by itself, and clearly no set of beliefs can warrant itself. Therefore, no such set can be in P’s warrant path. P, therefore, having a defective warrant path, is without warrant. Suppose, more particularly, that I wish to establish a certain visual law such as, "When appeared—to—red—bookly under conditions C, there is a high probability I am seeing a red book." SUPPOSG induction. 511 for in order t remember that 1 tally continuc this could all of the writte that only rev order for me 1 with the pres to rely on my the same recc be relying 01 am trying to round full c the class of observations 0f justifyi circle. SO eve still need ‘ how else cc tion, the j 59115 have t out r91ying of instifi, tional bel 205 book." Suppose that my method relies centrally on enumerative induction. Such a method also relies essentially on memory, for in order that the enumerations count as evidence, I must remember that previously recorded enumerations were experimen— tally continuous with the current one. One might respond that this could all be verified by reference to current observation of the written record of those previous enumerations. But that only reveals how memory dependent such evidence is. In order for me to know that those current records are continuous with the present enumeration, not rigged, forged, etc., I have to rely on my (or someone else’s) memory that these are indeed the same records made previously, etc. Moreover, I would also be relying on observation-—indeed, observation of the sort I am trying to justify—-and that simply brings the justification round full circle. In this case, P (or at least a member of the class of which P is a member, i.e., the class of Visual observations, each member of which also needs the same sort of justifying argument) is itself part of the justifying circle. So even if we simply ignored our reliance on memory, we still need to rely on observation to justify observation, for how else could we make the enumerative inductions. In addi— tion, the justifying value of enumerative induction would it- self have to be justified, and this could hardly be done with- out relying on both observation and memory. Thus, if the sort of justification BonJour seems to think is necessary for ra- tional belief actually is necessary, it seems highly unlikely that it can be as much in nee on beliefs Wh originally 5013 that one wouI justify obser ticular subse beliefs alone 15In a s argument migl belief of cl: formed from appeared—to-J visual appar is a belief {Z’} includ‘ whereas {Z} veridical vi Since a jus there will r tions of ty] "law of nat' But to just; to be anothe "laws" of t] i'e'l it Wi Surely that tion the 1' eventually Starting w: ant. .uuv that it can be offered without (a) using beliefs that are just as much in need of justification and (b) eventually relying on beliefs which are in the same class as that belief one originally sought to justify. It is also clear, incidentally, that one would be le no better position if she tried to justify observation beliefs in general as opposed to a par- ticular subset. Either way, justification by recourse to beliefs alone seems not at all feasible.15 15In a slightly more formal and generalized form, the argument might be put as follows. Let 0 be an observation belief of class Z where Z is defined as that class which is formed from observations of kind K. [E.g., O="I am being— appeared—to—red—bookly"; {Z}={o{o is a belief formed by human visual apparatus}. (Note that {Z} is a subset of {Z’}={ojo is a belief of the kind formed by human visual apparatus}. {Z’} includes hallucinatory and dream "visual" beliefs, whereas {Z} does not. But we are trying to justify only veridical visual beliefs, so we are concerned with {Z}.)] Since a justification of 0 will involve premises (1)—(3), there will need to be a premise to the effect that observa— tions of type 0 in conditions C are likely true, i.e., the "law of nature" In which (3) is will be used to justify 0. But to justify that particular premise (3), there will need to be another argument with a premise (3') to the effect that "laws" of the sort contained in (3) are generally veridical, i.e., it will require the justification of a "metalaw". So, e.g., "Laws of kind Ia are very likely true.“ But of course, that law, L2, will also need to be justified by some law, L“ and so on. We quickly come back upon a justification of some In by means of In, for clearly the justification of some Li will depend on observations, i.e., on the veridicality of {Z}. But then there will need to be a justificatory argument for {Z} which will include at least one empirical belief, and surely that empirical belief will require for its justifica- tion the justification of some 0 in {Z}, which, in turn, eventually will require the justification of {Z}. So even starting with a particular observation belief, we eventually come to the point. where 'the justification of 'that. belief requires the previous justification of the class of observa— tion beliefs of which it is a member, which in turn requires the justification of some of its members, and so on. Such circularity seems to run afoul of Plantinga’s criticism. Moreover, even if one were able to avoid this sort of circularity in the "premise (3) track", surely premise (2) requires the legitimating of an observation belief, and the It might justifies thes our other beli by BonJour are But (1) that d sibility of 0‘ they are to b: just “sees“ t justify that a thing? AI "intuition“ belief is ju other belief complicated The prc is the lack beliefs must there can he cannot get includes a1 beliefs bec their just 207 It might be replied that what in the final analysis justifies these observation beliefs is their coherence with our other beliefs and that arguments such as that schematized by BonJour are merely means by which that coherence is shown. But (1) that does not absolve those arguments from the respon— sibility of overcoming the difficulties pointed out above if they are to be valid arguments and (2) even if eventually one just "sees" that all these beliefs are coherent, one needs to justify that intuition. Just how is it that one "sees" such a thing? And what belief justifies the belief that that "intuition" is likely true, and then the belief that that belief is justified? Clearly, for BonJour, it must be some other belief, and again we either start a very large and complicated circle or proceed ad infinitum. The problem underlying the difficulties mentioned here is the lack of any reasonable stopping place. Since other beliefs must be used to justify any and all beliefs, logically there can be no stopping place. Under such restrictions, one cannot get outside the system of beliefs itself, for it includes all beliefs, even metabeliefs. So eventually all beliefs become justifications of beliefs which are used in their justification. The connection with reality which justifying of that premise will then get us on a route similar to, if not identical with, that described in the last para- graph. In short, since empirical justification of observation is of necessity going to employ observation, then observation will be justified by observation, which clearly is circular. The same can be said, mutatis mutandis for memory beliefs and authority beliefs. observation WE circularity 0f justify belie? either belief another belie: Justifyi seem to be : utilizing fou a very invol\ tional beliei . . . i claims V will th4 tions a appeal that I He is not c< an endeavor 208 observation was to provide coherentism cannot overcome the circularity of the justificatory process, for only beliefs can justify beliefs. But no stopping place is possible unless either beliefs can be justified by something other than another belief or there are basic beliefs. Justifying premise (2) [Conditions C obtain] would also seem to be a particularly difficult thing to do without utilizing foundational beliefs. Indeed, it is hard to see how a very involved circular regress would not result if founda- tional beliefs are not employed. As BonJour states, . . it is quite obvious that some of them [the claims which make up part of such a premise] . . . will themselves have the status of current observa- tions and hence will have to be justified . . . by appeal to the very same general sort of argument that I am presently considering. (SEK, 126) He is not concerned about the prima facia difficulties such an endeavor might present because "coherence enters in even at this level," observational beliefs "serving as justifica— tion for each other." (SEK, 126) Of course, it is clear that in a system of coherent observation beliefs, each will give support to the other, so in one sense, if I choose to justify one observation belief, utilizing a number of other observation beliefs to fill out premise (2) would seem entirely appropriate. BonJour, how— ever, feels that there would be no illegitimacy in so doing by means of a "round robin" set of justifying arguments wherein the observation beliefs {0} used in an argument to justify another observation belief B were themselves justified by arguments t Bx in a set C justified by 5 of observatior But while set, they cou supports my b desk, and yi belief that I of my belief time justify vicious cirC1 BonJour's as certainly no ity or size to support V but Planting The dif that only be needed to j my desk is fact that 1 that the 11 an argumen- Supports a] desk. The sort of im by arguments that utilized B. Thus, each observation belief B1:hi a set of coherent observation beliefs {0} would be justified by an argument in which premise (2) included a set of observation beliefs {{0} — Ba. But while each {{0} - 3d would support every other such set, they could not all warrant each other. Good lighting supports my belief that I am veridically seeing a book on my desk, and vice versa, but I cannot warrant or justify my belief that I am veridically seeing a book on my desk by means of my belief that the lighting is good, and then at the same time justify the latter by the former. Surely, that would be vicious circular argumentation plain and simple (contrary to BonJour’s assertion, SEK, 126), and that circularity would certainly not be made more palpable by reason of the complex- ity or size of the circle. Only if warrant were equivalent to support would such a justificatory scheme be acceptable, but Plantinga seems correct in claiming it is not. The difficulty here is compounded by BonJour’s insistence that only beliefs can justify or warrant beliefs, for what is needed to justify my belief that I veridically see a book on my desk is not a belief that the lighting is good, but the fact that the lighting is good. If all I have is a belief that the lighting is good, then that too must be supported by an argument. But if I have good lighting, then that fact supports and helps warrant my belief about a book lying on my desk. The obvious reply, of course, is that one never has the sort of immediate and "objective" access to "facts" that the previous senteI of direct reali ties, the prot realism is not cognizance of : by beliefs alo regress or 01'. able. But the or circular re of warrant, w Suppose, book on my de good (and, th item for a b lighting is r to that fact ing is good. some indepe' Contention. that. 1 mi< 5‘38 the boo? WOU1d seem, lngl I can 210 previous sentence supposes. But that reply is just a denial of direct realism. While direct realism has its own difficul— ties, the problem being pointed to here is that if direct realism is not in some cases possible, then all we would have cognizance of is beliefs, and beliefs will always be justified by beliefs alone. In that case, as stated above, an infinite regress or circular reasoning seems inevitable and unavoid- able. But then unless one can show how either such a regress or circular reasoning can be either the source or the conveyor of warrant, we are left with unjustified beliefs throughout. Suppose, for example, I say that my belief that I see a book on my desk is warranted (partly) because the lighting is good (and, therefore, presumably I am not mistaking some other item for a book). Then I must either claim that I know the lighting is good because I do have some sort of direct access to that fact or I must simply claim that I believe the light— ing is good. But if I choose the latter, surely I must have some independent reason (belief or fact) supporting' that contention. I can’t simply point to the book as proof of that. I might, of course, point to the clarity with which I see the book, but that is a different belief altogether. It would seem, moreover, that if I am to avoid circular reason- ing, I can never point to my seeing the book itself as proof that the lighting is good. Since the evidence that the lighting is good is bound to include other observation be- liefs, and they, in turn, can never use the belief that the lighting is good in their substantiation, we are bound to go on forever unll or foundationa In other foundational 1 as necessary f cannot be vali of observati01 beliefs. In discu argues that t The “Doxastic tation of 111 correct." ( could be sai one implicat ing a sort assume this must itself re