II M J” I l ' JIM/WI! ”Will! I“ J WWW 5.- FGTWMEQFY (355' E73! 585555313: ”555; ASWG‘F ”EMF? 145 213 THS flies-rs {:69 ‘f‘fm ”Degree 05 54.15%. ITCRELJ‘H 3‘5 355:". UHEE E55. SELFY S”. Cfiwies Lazar 2%9 amt/Ht! ABSTRACT TOWARD A FOR MALIZA TION OF DEMOGRAPHIC TRANSITION THEORY by S. Charles Lazer An explication and analysis of demographic transition theory yield sixteen propositions which are adduced to account for the phenomena described by the theory. These propositions form the basis of a planned computer simulation of demographic transition. Setting the parameters of this model according to historical data make it capable of partial verification as a theory, while changing parameters and processes lend the model the ability to explore the theory more fully. TOWARD A FORMALIZA TION OF DEMOGRAPHIC TRANSITION THEORY by S. Charles Lazer A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology 1969 ACKNOWLEDGMENTS The author would like to thank the many people, faculty and graduate students in the Sociology Department of Michigan State University, whose support and assistance facilitated the completion of this work, especially Professors Harvey Choldin, John T. Gullahorn, and Grafton D. Trout whose help, individually and in committee was invaluable. Thanks are also due to Beth Benson and Isabelle Hamm for their patience in typing the several drafts of the thesis and to Kathy McKinney whose enlightening and enchanting discussions were of tremendous stimulation and value. ii TABLE OF CONTENTS Toward a Formalization of Demographic Transition Theory ..................................... 1 Bibliography .......................................... 32 Appendices Appendix A: Inventory of Variables Considered to affect the Mortality Rate .......... 36 Appendix B: Inventory of Variables Considered to affect the Fertility Rate .......... 433 Table 1: Table 2: Table A1: Table A2: Table A3: Table A4: Table A5: Table A6: Table A7: Table Bl: Table B2: Table B3: , LIST OF TABLES Variables Influencing the Mortality Rate (MORTY). . . . . ........ . . ............... Variables Influencing the Fertility Rate (FERTY) ............ . ................. Variables Considered to Affect or Subsumed by "Political stability" (ORDER). . . . .......... Variables Considered to Affect of Subsumed by I”Improved resource extraction" (AGTEX) ...... Variables Considered to Affect or Subsumed by "Improved food supply and diet" (DIET) ........ Variables Considered to Affect or Subsumed by "Improved transportation and communication network" (TRANCOM) ....................... Variables Considered to Affect or Subsumed by "Improved public sanitationH (SANTEX) ........ Variables Considered to Affect or Subsumed by ”Increased medical technology and improved medical care" (MEDTEX) .................... Variables Considered to Affect or Subsumed by "Modernization" (MOD) ...................... Variables Considered to Affect or Subsumed by ”Contraceptive methods, their accessibility and availability" (OZ) ........................ Variables Considered to Affect or Subsumed by "Traditional value system predisposed to high fertility" (CUSTOM) ........................ Variables Considered to Affect or Subsumed by "Urbanization" (URBAN) ...................... iv 14 17 36 37 38 39 40 41 42 43 44 45 Table B4: Variables Considered to Affect or Subsumed by ”Industrialization" (INDUST) ................. 46 Table B5: Variables Considered to Affect or Subsumed by "Modernization" (MOD). .................... 47> Figure 1: Figure 2: Figure 3: Figure 4: LIST OF FIGURES The Demographic Transition ................. Plausible Vital Rate Cycles .................. Mortality Decline. '. ......................... Fertility Decline ............................ vi 7 20‘ 2.3 It is difficult to avoid the conclusion that the major advances in the science of population have come from improvements in the sources of information and in the techniques of analysis rather than from the broad interpretations. 1 The lack of valid formal theory in demography has been re- 2 peatedly indicated and the need for it is sharply felt. This lack is especially striking, if only for the fact that "over-arching theories. . . . .3 . . (have) dominated so much of the nineteenth century work ' 1n the dis- cipline, without a commensurate advance in the development of theory. At best, what are generally considered by demographers to be theory are empirical generalizations, i. e. , "a set of uniform conjunctions of traits repeatedly observed to exist, without any understanding of why 4 the conjunction occurs; without a theory which states its rationale. " At worst, the preoccupation with such "over-arching theories" which 1. Kingsley Davis, "The Sociology of Demographic Behavior, " in Robert K. Merton, Leonard Broom, Leonard S. Cottrell, Jr. (eds. )5 Sociology Today (New York: Basic Books, 195 9), p. 313. 2. See for example, Robert Gutman, "In Defense of Population Theory, " American Sociological Review, v. 25, 1969, pp. 325-333 and Rupert Vance, "Is Theory for Demographers, " Social Forces, v. 31, 1952, pp. 9-13. 3. Frank Notestein, in B. F. Haley (ed; )l. SUPVBX 9_f_ @gntemgonang Economics, v. II. (Homewood: Irwin, 1952), p. 129, cited in Warren C. Robinson, ”The Development of Modern Population Theory, " American Journal o_f Economics and Sociology, v. 23, 1964, p. 376. 4. John Dewey, Logic: TfhewTheDry gflfigguh'deew'Yorkf. Holt',';_. 1938), p. 193. are often no more than simply "a long stretch of verbal analysis" has probably retarded the development of the discipline. 6 One of the dominant theories, if we understand theory in demography to be used in its broadest sense as a set of general state- ments or principles which ”organize existing knowledge, lead to the acquisition of new knowledge, and help in the solution of population problems, "7 today is that of the demographic transition, an attempt to explain population growth and which is derived historically from the vital experiences of western Europe. It would scarcely be an exaggeration to call it "The" modern population theory. . . . (however). . . . It is at the same time a very general theory lacking in rigourous formulations, or neat precise answers to complex questions of population change.8 9 Since its initial formulation by Thompson in 1929 some at- tempts have been made to state the theory in a more formal manner. 10 5. Kingsley Davis, loc.cit. 6. Frank Notestein, loc. cit. 7. Robert Gutman, 92. cit. , p. 333. 8. Warren C. Robinson, 22.91. p. 389. 9. Warren S. Thompson, "Population, American Journal 93‘ Sociology, v. 34, 1929, pp. 959- 975. 10. See for example, Donald 0. Cowgill, "Transition Theory as General Population Theory", Social Forces, v. 41, 1962, pp. 270 - 274 and Paul K. Hatt, N. L. Farr, E. Weinstein, ”Types of Popula- tion Balance," American Sociological Review, v.20, 1955, pp. 14- 20. 3 However, "the reasoning underlying the statement remains implicit at the present time. "11 This paper, then, will attempt to present the transition theory in systematic fashion, hopefully stating the as— sumptions and propositions explicitly in terms of the proposed simu- lation in order to move from the level of description to that of ex- planation. Such a move can only facilitate the determination of a causal nexus and its underlying rationale. To state that there is some confusion in Sociology as to the meaning of the term and the nature of ”Theory" is an understatement, and the realm of this question is beyond the scope of this paper. How- ever, to state categorically, as some have done, that Grand programs for constructing theories have not enlarged our knowledge and understanding of society nor have they proved to be realiable guides for social research 12 and further that such theories are by their definition devoid of em— pirical referents can only be at best misleading. Theory, or more specifically, logical explanation can be de- fined as consisting of two parts, the explanandum and the explanans, such that By the explanandum, we understand the sentence describing the phenomenon to be explained (not that phenomenon itself); by the explanans, the class of those sentences which are 11. Robert Gutman, pp. cit. , p. 332. 12. Ibid. , p.333. 4 adduced to account for the phenomenon. . . . the explanans falls into two sub-classes; one of these contains certain sentences C1, C2, . . . . Ck which state specific antecedent conditions; the other is a set of sentences L1, L2, . . . . Lr which represent general laws. 13 Further, the explanation must meet certain criteria of adequacy: 1) The explanandum must be a logical consequence of the ex- planans, in other words, the explanandum must be logically deducible from the information contained in the explanans. . . 2) The explanans must contain general laws, and these must actually be required for the derivation of the explanandum. . . 3) The explanans must have empirical content, 1. e. it must be capable, at least in principle, of test by experiment or ob- servation. . . 1 4) The sentences constituting the explanans must be true. This, then, is "Theory" but what of the "Demographic Transi- tion”? The transition) is an attempt to explain the rapid population growth experienced by the "Western" civilizations in the recent past and the ensuing S-curve depicting population size, and thence to pre- dict the growth which could be expected in the developing nations as they approach and then pass through successive stages of moderniza— tion. In the first statement of the transition, selected countries of the world were classified according to their natural rates and 13. Carl G. Hempel and Paul Oppenheim, "The Logic of Explanation" in Herbert Feigl and May Brodbeck (eds. ), Readings_i£ the Philosophy 31" Science (New York: Appleton-Century-Crofts, 1953,) p. 321. 14. Ibid, pp. 321-322. 5 consequently their growth rates, with the processes underlying the taxonomy left implicit. The classifications arrived at were: below: Group A countries: Very rapid declining birth-rate and death- rate with the former declining more rapidly than the latter so that the rate of natural increase is also declining. Group B: . . . . decline in both birth- rates and death-rates is underway in certain classes, but. . . . the death-rate is declin- ing rapidly or even more rapidly than the birth—rate with the result that the rate of natural increase will probably for some- time remain as great as now, or even become larger in the near future. Group C: Both birth-rates and death-rates are less controlled than in either A or B. But in some of these countries. . . . there is some indication that death-rates are coming under control faster than birth-rates. In such of these lands as are develop- ing modern industry and sanitation, there is very likely to be a very rapid increase in numbers during the next few decades. In many of these lands, however, both birth-rates and death- rates are quite uncontrolled and we may expect either a rapid increase or almost a stationary population dependent upon the harshness of the "positive” checks to population growth. . . . 5 Represented graphically, the three groups would appear as 11 A BIRTH RAY! I DEATH RAY! I I I I I l I I I J FIGURE 1: The Demographic Transition 15. Warren S. Thompson, 92.21;. , pp. 961-962. 6 The countries classified as belonging to Group A were the most modern, those of Group B were undergoing modernization, and those in Group C were described as primitive. The representation is one of a shifting control over the natural functions--from a biological determi— nism of mortality and fertility to a more rational one. 16 Rapid growth occurs because death rates tend to fall more quickly than do birth r ites. The first statement of what has since become known as the demographic transition is simply: Declining rates of mortality and fertility are asso- ciated with modernization, and further that mortality will decline earlier than fertility. Although there are at least four simple cycles of the vital rates that will yield the S-shaped demi-cycle of population growth experienced by the nations which have undergone modernization17 (see Figure 2 be- low) only Cycle II is associated with transition theory. Cycle I may be found in primitive societies where fertility is relatively fixed and growth is dependent solely on fluctuations in mortality. Cycle III has not mani— fested itself as yet, but may appear in future societies wherein mortality control can be extended no further so that growth may only be determined by fluctuations in fertility. Cycle IV has not appeared to this date, and there is little reason to assume that it will appear in the future. 16. See, for a brief discussion, Leighton van Nort, "Biology, Rationality, and Fertility, a Footnote to Population Theory, " Eugenics Quarterly, v. 3, 1956,pp:.‘.157_-1«60.n+mn is. 17. Donald 0. Cowgill, "The Theory of Population Growth Cycles, " American Journal o_f Sociology, v. 55, 1949, pp. 163-170. BIRTH RATE DEATH RATE DEATH RATE CYCLE I CYCLE II BIRTH RATE BIRTH RATE DEATH RATE but" am CYCLE III CYCLE IV FIGURE 2: Plausible Vital Rate Cycles Consequently, further development of the theory-—mostly 18 Kingsley Davis, 19 and especially notably by A. M. Carr-Saunders, Frank Notesteinzouwas concerned primarily with the establishment and refinement of its implicit causal mechanisms and not with the 18. A. M.Carr-Saunders, World Population (Oxford: Clarendon Press, 1936). 19. Kingsley Davis, "The World Demographic Transition, " The Annals, v.237, 1945, pp. 1-11. ‘ 20. Most notably, Frank Notestein, "Population - The Long ViewH in Theordore W. Schultz (ed. ). Food for the World (Chicago: University of Chicago Press, 1945), pp. 36-57. 8 nature and shape of the growth pattern as postulated. 21 A succinct modern statement of the sequence of events underlying the demcr graphic Wihm may be given as follows: The agrarian low-income economy is characterized by high birth and death rates--the birth rates relatively stable and the death rates fluctuating in response to varying fortunes. Then as the economy changes its form to a more interdepen- dent and specialized market-dominated economy, the aver- age death rate declines. It continues to decline under the impact of better organization and improving medical know- ledge and care. Somewhat later the birth rate begins to fall. The tworates pursue aI-more' ”or. less parallel downward course with the decline in the birth rate lagging behind. Finally, as further reductions in the death rate become harder to obtain, the birth rate again approaches equality with the death rate and a more gradual rate of growth is reestablished, with, however,low risks of mortality and small families as the typical pattern. Mortality rates are now relatively stable from year to year and birth rates-mow responsive to volun— tary decisions rather than to deeply imbedded customs--may fluctuate from year to year. 22 We now have a verbal statement of some of the relationships implicit in earlier formulations. The variables incorporated into this model as influencing mortality and fertility are market economy, organization, and medical care. However, because of the difficulty of extracting precise 21. The scope of this paper precludes a detailed dicussion of the evolution of the Demographic Transition Theory. However, for an extensive discussion of the matter and an almost exhaustive English bibliography on the subject, see Grafton D. Trout. Jr. , "The Develop- ment and Current Status of the Demographic Transition Theory, A Bibliographic Survey" (1963, unpublished). 22. Ansley J. Goals and Edgar M. Hoover, Population Growth 9.9.9 Economic Development _in_ Low-Income Countries (Princeton: Princeton University_Press, 1958), p. 13. 9 information from such verbalizations, some recent attempts have been made to state the theory in propositional form. The advant- ages of this form are that "it forces him (the theorist) to spell out his assumptions, to make explicit his deductions, and it will remind him of any bypassed implications. "23 Hatt, Farr and Weinstein posit That mortality and fertility are so related to urbanization and industrialization that low levels of the vital rates are associated with high levels of modernization; and that high levels of vital rates are associated with low levels of modernization; and further, that medium levels of moderni- zation will serve to depress mortality more rapidly than fertility. 24 The propositions are formulated in this fashion in order that some of the relationships expressed in the theory may be assessed. But it is interesting to note that a causal connection in only brought out in the third part of the statement. A second formal statement of some aspects of the transition theory is somewhat more ambitious. Shifting "from the use of the economists' conventional variables-—land, labor and capital-—to population, technology and culture','25 Cowgill sets forth twelve pro- positions which are in essence, generalizations induced from 23. Hans L. Zetterberg, 9g Theory and Verification i_n Sociology. Third Enlarged Edition (Totawa, New Jersey: Bedminster Press, 1965). P. 100. 24. Paul K. Hatt gt §_l_. , 92. cit. , p. 15. 25. Donald 0. Cowgill, 22%,. (1'Nfl}.p@MpS ("It 5”. :3; I'-:I'i-i:-., Vat ,I'OZ‘i‘L 10 empirical research in population growth, and which are more con- cerned with various differential rates experienced by populations in transition from high equilibrium mortality and fertility to low equili- brium mortality and fertility, than with the generation of the transition itself. below: 1) 2) 3) 4) 5) 6) The propositions relevant to the transition itself are set down In the absence of effective technology of birth control and death control, population will increase to a maximum carrying capacity of the environment, then achieve a stable equilibrium characterized by high birth rates and limited longevity. Under conditions of industrialization and urbanization, given the technology of birth control and death control, there is a marked tendency for the technology of death control to be applied earlier and more extensively, resul- ting in rapid population growth and extensive structural changes in the population. Later, and less predictably, cultural values permitting, the technology of birth control may be employed to reduce the rate of growth and bring about a new condition of equilibrium. During the process of urbanization and industrialization, the nuclear family tends to replace extended and consan— guine forms. With a falling birth rate, the size of the nuclear family tends to decline. During the demographic transition there is a marked shift from extractive industries and agrarian occupations to industrial and commercial occupations. During the demographic transition there is a marked tendency toward urbanization of the population. 26 26. Ibid. 11 The need for a formal statement of the demographic transia tion theory still exists. According to Cowgill, the most pressing need is for those studies which will shed the most light on "the factors in- hibiting and promoting the application of birth control technology. "27 However, that the need for increased knowledge ofthose factors af— fecting mortality decline as well is no less pressing has been indi— cated in not so gentle a manner by some critics of the Theory. 28 In other words the question still asked is: ”How does transition theory -. work? ", both in terms of historial explanation of the European ex- perience and of prediction in the ThirtiuWSchIdmdday. In order to gain the greatest amount of understanding of the demographic transition, and to facilitate the formulation of proposi— tions regarding it, the transition was considered to consist of two processes: mortality decline and fertility decline. This is not to imply that the two are unrelated, but simply that the greatest benefits can be obtained by treating them separately, without doing serious damage to the Theony... The factors which affect one do not always affect the other and,’ in historical accounts of the transition wherein 27. Ibid, p.274. 28. See, for example, William Petersen, "The Demographic Tran- sition in the Netherlands, " American Sociological Review, v. 25, 1960, pp. 334-347, and Leighton van Nort, "Some Issues for Tran- sition Theory, " Population Index, v. 26, 1960, pp. 387-395, and Dennis Wrong, Popglation (New York: Random House, 1956), Ch. 2. 12 variables considered to affect the vital rates are endogenous rather than exogenous, do not occur simultaneously. In addition, the vari- ables considered to influence the vital rates were considered to act as independently of each other, not because their interdependence was felt to be either non—existent or negligible, but rather because it was felt that this would introduce severe complications into the model, which could only render it almost unworkable at this early stage of the analysis. In Zetterberg's terms, "propositions with two variates are acceptable as intermediary steps in theory construction even if they do not tell the whole story. Once formulated they lend them- selves to amendments. "29 These amendments are logical and neces— sary refinements which the model must undergo, but the independence restriction is necessary at this point. The dependent variables of the propositions--statements of relationships between at least two variable330-—formulated below were considered to be a) the mortality rate and b) the fertility rate. The independent variables were derived from the works of several students of the problem. The variables listed in Table 1 are superordinate categories of the generally more specific variables 29. Hans L. Zetterberg, 22. cit. , p. 65. 30. Ibid, p.64. 13 indicated as affecting the mortality rate. The names used to repre— sent the variables in the model are given in parentheses. l4 X X X X X X X 5250: use 3.30 X X X comacucm x x x 59.60 x x x x x x 323 X X X X X X X aoxoafim X. X X X X X X misc X X X X X X 53?.qu X X X X X mauvgamuhhau . X X X econ—ESE. AXMHDMEV AEOUZoaa5~ “50.33:: po>osa5~ 695.23: 30330“ A>HKOSC wh.:4<.HmO—2 writ OZ—UZMDJrAZH wfldfl<$~<> a aunt—mgr l5 Propositions concerning the declining mortality rates des- cribed by transition theory were then formulated using the variables shown in Table 1 and which were felt to accurately represent the re- lationships implied or expressed by the theory. The variable "modernization" was not used inasmuch as it was felt that the other variables were all manifestations of it. ”Modernization" is differen- tiable from the other variables if it is conceived of in terms of an attitudinal configuration, rather than technological progress. How- ever, because mortality reduction is almost universally viewed posi- tively, it is unnecessary to take it into consideration in the model being developed here. If a) A primitive state of no significant mortality control will serve to maintain a high mortality rate is initially posited the following propositions regarding the mortality rate in the demographic transition may then be generated: b) Increased political stability will contribute to a declining mortality rate. 0) Improved resource extraction will contribute to an impro- ved food supply and diet. (1) An improved transportation and communication network will initially contribute to an improved food supply and diet. e) An improved food supply and diet will contribute to a 16 declining mortality rate. f) Improved public sanitation techniques will contribute to a declining mortality rate. g) Increased medical technology and improved medical care will contribute to a declining mortality rate. h) Continued declining mortality will eventually yield (when the limit of mortality control is achieved) a low stable mortality rate. By the identical process the variables considered to influence the fertility rate as shown in Table 2 were induced. I7 X X X X w. .5353— .323an X .325: can 380 .82 cm.» £39.33 .uash .33.— xh v— s 3. 02m .. 32.5 Souuofivz X nacvgamLhaU X coma—Bosh ADA<> Dzv 55:8 30— 3 ponamzpoaa 53me os~m> ADO—2v catas_caooo§ AFmDozc COSmNZEaumspfi AzF~4_Hmmh ”In. UZ—UZHDJKZH WMAQSG<> N MJQ NUVALU .9 _> __> _> NUVALU URBAN DEFERT DEF ERT TRA NCOM 9 URBAN (a) (b) (c) (d) (e) (f) (g) (i) (j) (I?) (l) (m) (n) (o) (1)) Thus a system of propositions has been created from which the demographic transition phenomenon may be deduced. The propositions all contain empirical content, they can all be verified, and if it is as- sumed that the propositions may achieve the status of general laws, then the system is capable of verification as a logical explanation or theory according to the criteria of Hempel and Oppenheim cited above. With demographic transition theory stated in propositional 21 form, the construction of a rudimentary stochastic simulation model is greatly facilitated. One need only make the further stipulation that the variables act sequentially as well as independently. That this as- pect of the transition is not unwarranted in any event is implied by descriptions of populations which have naturally undergone the transi- tion from high mortality and fertility to low mortality and fertility. 31 Flow charts of the simulation models of the mortality and fertility declines are shown in Figures 3 and 4 below and it is interes— ting to note that proposition 15) TRANCOM --> URBAN permits the connection of the two simulations, and at the same time introduces the element of a time lag between the initial decrease of mortality and that of fertility. 31. See for example, Kingsley Davis, gp.gi_t_. (1945), Frank Note- stein, gp.cit. (1945), and Dennis Wrong, pp._c_i_t_. .4 YES nascent >d? NO FIGURE 3: MORTALITY DECLINE 4 GENERATE MTV YES ORDER ‘ L P so A A i ' INCREASE YE: GENERATE DIET A07“ no N MTV 3! DECREASE VALUE 0, MTV; L scan “'7 A M l3, I2 L< MTV Vob. As with the probabilities of occurrence, Voi may be derived either historically or from a set of propositions regarding the population being simulated. Voi is then evaluated at cell 3, the point "ORDER>a? ”, where a is assumed to be the minimum value of Voi which will affect the mortality rate. Consequently, if Voiga the model loops back to ”Generate ORDER" and generates Voi+1.: This loop is repeated 1+n >a, at which point the death rate can be decreased at ”Decrease MORTY " (cell 4). The magnitude of a is obviously related to Voi and is determined in the same fashion. The operation "Decrease MORTY" is actually a two-stage process. Continuing with the example, ORDER, Voi may be cons- trued to be the impact of Oi on the mortality rate. Consequently, it may be assumed that the amount of mortality decline generated in "Decrease MORTY" is proportional to Voi - a, where Voi>a. The exact nature of the proportion can be deduced from historical data, or from a set of propositions concerning the simulated population. The repeated mention of more than one method of assigning values to the parameters of the model arises from the multiplicity of functions which the model is capable of performing. Its major functions would be partial verification of Hemmgmhm' transition:- 28 theory and prediction of population growth under varying conditions, as well as exploration of the theory itself. The most pressing theore- tical question historically is that of the nature of the relationship bet- ween the components of the variable "Modernization" and its effect on the social structure and value system. With regard to the present, an examination of the effects on the process of changes in the order of the events as well as extensions and compressions of the time factor in— volved would almost certainly prove fruitful. However, before either operation could be attempted with any degree of confidence in the relevance of the findings,the model itself must be validated. The validation technique employed would be that of finding a "best-fit simulation" in a manner similar to that followed by Pool, 32.511. , in Candidates, Issues and Strategies: AComputer Simulation. 32 The data on vital rates for several Western European countries having undergone the demographic transition "naturally" is available and is considered reasonably accurate as far back as the eighteenth century. Calibration would then be concerned with adjusting the processes and parameters of the model so that the output of the model most closely approximates the vital experience of the actual population it is being compared with. This "best-fit" technique would not only attempt to achieve a correspondence at the beginning and end of the transition, but also to achieve a correspondence at each stage 32. Ithiel Pool, R. P.Abelson and Samfiel Popkin,Candid§gtesJalss_ues and Strategies: AComputer Simulation (Cambridge: The MIT Press, 1965). ,A 2 in the process. This is necessary because similar outputs of the simulation and the actual experience do not necessarily imply that the processes by which the outcomes are determined are congruent. If the outcomes for the several stages of the vital processes and of the model are also similar the probability of the model's being an accurate representation of vital experience is greatly increased. Once the model has been calibrated and the parameters estab- lished the use of the model for theory verification is straightforward. Renewed interest in historical demography has made itself felt in re- 33 and has improved the quality as well as quanitity of data cent years available on populations of two centuries ago. Furthermore, techni— ques of analysis of such data have also been improved. Consequently, a more accurate portrait of historical population growth can be pre- sented. If the simulation is constructed with inputs identical to those of history, then the outcomes may be compared. Congruent outcomes will not prove the theory or any of its propositions true, but dis- similar outcomes are certainly capable of proving the model invalid and seriously questioning the value of the theory. That the model is at this point rudimentary cannot be denied. However, its form makes it readily amenable to refinement. The major refinement, which should be made as soon as possible, is the 33. See, for example, Daedalus, v.97, 1968, and David Victor Glass and D. E. C. Eversley (eds. ), Population i_n History: Essays i_n Histori- cal Demography (London: Edward Arnold, 1965). 30' establishment of an interdependence between the variables posited as decreasing mortality and fertility. At this point in time, there are only two indications of even the simplest interdependence displayed in the simulation model. These simple feedbacks occur in the DIET- TRANCOM cycle and the URBAN-INDUST cycle. Although it was felt that at this time it would be more valuable to consider all variables as dependent, independent or intervening, it is almost inconceivable that there are no relationships between interdependent variables in the scheme. This discrepancy should be minimized as soon as it is possible to do so, in order to improve the descriptive as well as pre— dictive accuracy of the model. The most powerful tool in determining the answers to the questions of interdependence, sequentiality, and time-span may well be the model itself. Inasmuch as these are empirical questions, the simulation has the advantage of reality if only because reality cannot be re-run, as can a computer program. Additionally, by its very structure the model is capable of undergoing shifts in levels of anal- ysis. Because the variables selected for use in the model are super- ordinate classes of other variables, the transformation into a model of less abstraction such as the one put forward by Davis and Blake34 in 1956 is straightforward. It is» also not difficult to conceive of the 34. Davis, Kingsley and Blake, Judith, "Social Structure and Fertility: An Analytical Framework, " Economic Development and Cultural Change, v. 4, 1956, pp. 211-235. 31 formulation of the model in the more abstract terms of Duncan's ecologi— cal complex. 35 Seventeen years ago, the president of the Population Association of America indicated the need for the development of demographic theory and assured members of the Association that even ”he who develops a theory capable of being proved invalid makes a contribution. ”3 The formal statement of transition theory herein, and the stochastic model which is its extension, are together an attempt to help remedy the situ- ation. If the model is presumed to be an accurate representation of actual mortality and fertility declines and consequent population growth, then it can be used to predict the effects on the vital rates of changes in the independent and intervening variables. At this time, when the question of overpopulation is of crucial importance to the survival of mankind, the ability to predict the consequences for growth rates of innovations, be they technological, spiritual, or organizationallsis of the‘utmost importance, and this ability would be no minor factor contributing to humanity's continued existence. 35. Otis Dudley Duncan, "From Social Sytem to Ecosystem", Sociological Inquiry, v. 31, pp. 140—149. 35. Ruper Vance, pp.cit. BIBLIOGRAPHY 10. 11. 12. 13. 14. 32 BIBLIOGRAPHY Blacker, C. P. , "Stages in Population Growth, " Eugenics Review, v.39, 1947, pp. 88-102. Carr-Saunders, A. M. , World Population " (Oxford: Oxford University Press, 1936). Coale, Ansley J. and Hoover, Edgar M. , Population Growth and Economic Development _i_n_ Low-Income Countries (Prince- ton: Princeton University Press, 1958). Cowgill, Donald 0. , "The Theory of Population Cycles, ” American Journal 2f Sociology, v. 55, 1949, pp. 163—170. Cowgill, Donald O. , "Transition Theory as General Population Theory, " Social Forces, v. 41, 1962, pp. 270—274. Daedalus, v. 97, 1968. Davis, Joseph S. , "Population and Resources: A Discussion of Papers by Frank W. Notestein and P. V. Cardon, "Journal o_f E American Statistical Association, v. 45, 1950, pp. 346-349. Davis, Kingsley, Human Society (New York: Macmillan, 1949). Davis, Kingsley, "The Sociology of Demographic Behavior" in Robert K. Merton, Leonard Broom and Leonard. SI. Cottrell (eds. ) Sociology Today (New York: Basic Books, 1959), pp. 309-333. Davis, Kingsley, "The World Demographic Transition, " The Annals, v. 237, 1945, pp.1-11. Davis, Kingsley and Blake, Judith, "Social Structure and Fertility: An Analytical Framework, "Economic Development and Cultural Charng v. 4, 1956, pp. 211-235. Dewey, John, Logic: The Theory 9f Inguiry (New York: Holt, 1938). Duncan, Otis Dudley, "From Social System to Ecosystem,H Sociological Inquipy, v. 31, pp. 140-149. Durand, John D. , "Generalized Demographic Models and Projections of the Population of China," Asian Survey, v. 1 no. 4, 1961, pp.29-34. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 33 Glass, David Victor and Eversley, D. E. C. (eds. ). qulation jp History: Essays i_n Historical Demography (London: Edward Arnold, 1965). Gullahorn, John T. and Gullahorn, Jeanne E. , "The Computer as a Tool for Theory Development", in Dell Hymes (ed. ), jig I_I_sp o_f Computers _ip Anthropology. (The Hague: Mouton, 1965). pp.427-448. Gutman, Robert, "In Defense of Population Theory, " American Sociological Review, v. 25, 1960, pp. 325-333. Hatt, Paul K., Farr, Nellie L., and Weinstein, Eugene, ”Types of Population Balance," American Sociological Review, v. 20, 1955, pp.14—20. Hauser, Philip M. and Duncan, Otis Dudley, "Overview and Conclusions, " in The Studygi; Population: An Inventopy and Appraisal (Chicago: University of Chicago Press, 1959), pp. 11-26. Hempel} Carl G. , and Oppenheim, Paul, "The Logic of Explanation, " in Herbert Feigl and May Brodbeck (edsil)Readings i_n_ the Philosophy 9f Science (New York: Appleton-Century—Crofts, 1953) pp. 319-352. Kemeny, John G. , Snell, J. Laurie, and Thompson, Gerald L. , Introduction 19 Finite Mathematics (Englewood Cliffs: Prentice— Hall, 1956). Kirk, Dudley, "Dynamics of Human Populations, " Eugenics Quarterly, v.2, 1955, pp. 18-25. Langer, William L. , "Europe's Initial Population Explosion, " in David M. Heer (ed. ), "Readings on Population (Englewood Cliffs: Prentice-Hall, 1968), pp.2-15. McKeown, Thomas and Brown, R. G. , "Medical Evidence Re— lated to English Population Changes in the Eighteenth Century, " David M. Heer,(edsr)IReadings .ngngulatigp.(Englewood Cliffs: Prentice-Hall, 1968), pp. 16—38. Notestein, Frank, "The Population of the World in the Year 2000, " Journal o_f the American Statistical Association, v.45, 1950, pp. 335-345. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 34 Notestein, Frank, "Population--the Long View, " in Theodore W. Schultz (ed. ), Food for the World (Chicago: University of Chicago Press, 1945), pp.36-57. Petersen, William, Population (New York: MacMillan, 1961). Petersen, William, "The Demographic Transition in the Netherlands, " American Sociological Review, v. 25, 1960, pp.334-347. Pool, Ithiel, Abelson, R.P. and Popkin, Samuel, Candidates, Issues gStrategies: A Computer Simulation (Cambridge: The MIT Press, 1965). Robinson, Warren C. , "The Development Of Modern Population Theory, " American Journal 5% Economics and Sociology, v. 23, 1964, pp.375-392. Stolnitz, George J. , "Interrelations Between Economic Develop- ment, Levels of Living and Demographic Trends, " in Bogue, Donald J. , (ved.), Applications o_f Demography: The Pgulation Foundation, 1957), pp.5-14. Thompson, Warren S. , Population Problemsj Fourth Edition, (New York: McGraw-Hill, 1953). Thompson, Warren S. , "Population, " Sociology, v.34, 1929, pp. 959-975. A meric an Journal _o___f Trout, Grafton, D. , Jr. , "The Development and Current Status of the Demographic Transition Theory, A Bibliographic Survey" (1963, unpublished). Vance, Rupert B. , "Is Theory for Demographers,‘ Social Forces, v.31, 1952, pp. 9-13. van Nort, Leighton, "Biology, Rationality and Fertility, A Footnote to Transition Theory, " Eugenics Qparterly, v.3, 1956, pp. 157-160. van Nort, Leighton, "Some Issues for Transition Theory, " Population Index, v. 26, 1960, pp. 387-395. van. Nort, Leighton and Karon, Bertram P. , "Demographic Transition Re-examined, " American Sociological Review, v. 20, 1955, pp. 523-527. 39. 40. 41. 42. 35 Whelpton, P. K. and Kiser, Clyde V. ,"Trends Determinants, and Control in Human Fertility, " The Annals, v. 237, 1945, pp.112-122. Wrong, Dennis, Population (New York: Random House, 1956). Yaukey, David, "On Theorizing about Fertility, " American Sociologist, v.4, 1969, pp. 100-104. Zetterberg, Hans L. , 9;; Theory and Verification i_n Sociology, Third Enlarged Edition (Totawa, New Jersey: Bedminster Press, 1965). APPENDICES APPENDIX A INVENTORY OF VARIABLES CONSIDERED TO AFFECT THE MORTALITY RATE 36 TABLE Al VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY "POLITICAL STABILITY" (ORDER) Author Factor Carr-Saunders Political Security Notestein Domestic Order Davis Democratic Institutions Blacker General Security Strong Central Political Control Wrong Political Unity Coale and Hoover Political Organization 32‘ TABLE A2 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HIMPROVED RESOURCE EXTRACTIONH (AGTEX) Author Factor Thompson Improvement of agricultural techniques Carr—Saunders Food production Notestein ~ Improved farming methods The New World and its resources Davis Improved agricultural techniques Feasible commercial agriculture New World resources Agriculturalism Blacker Improved agricultural techniques Wrong Agricultural techniques Cowgill Mechanized agriculture Viable commercial trade areas Coale and Hoover Market economy «2'8 TABLE A3 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HIMPROVED FOOD SUPPLY AND DIET'I (DIET) Author Factor Carr-Saunders Improved diet Notestein Available food supply Davis Regular, more abundant, food supply Blacker Regular, more abundant food supply Wrong Improved diet Petersen Improved diet Coale and Hoover Improved diet 89 TABLE A4 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HIMPROVED TRANSPORTATION AND COMMUNICATION NETWORK” (TRANCOM) Author Factor Davis Improved transportation network Blacker Improved transportation network Wrong Improved transportation network Coale and Hoover Improved transportation - communi— cation network 40' TABLE A5 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HIMPROVED PUBLIC SANITA TIONH (SANTEX) Author Factor Thompson Modern sanitation Carr-Saunders Sanitation techniques Notestein Sanitation Davis Public sanitation Blacker Elementary sanitation techniques Wrong Sanitation and public health Petersen Amelioration of environment Coale and Hoover Improved sanitation techniques 4 1' TABLE A6 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HINCREASED MEDICAL TECHNOLOGY AND IMPROVED MEDICAL CARE" (MEDTEX) Author Factor Carr-Saunders Modern medical techniques Notestein Modern medical techniques Disease control Davis Scientific medical techniques Blacker Modern medical services Wrong Modern medical techniques Cowgill Scientific medical practice Petersen Change in balance between infecting organism and host Preventive or curative therapy Coale and Hoover Improved medical techniques 42 TABLE A7 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY "MODERNIZATION" Author Thompson Carr — Saunders Notestein Davis Blacker C ow gill Coale and Hoover (MODERN) Factor Industrialization Clothing Industrialization Productivity Modernization Level of living Machinery Scientific ideals Humanitarian sentiments Industrialism Regular source of fuel Machinery Industrialization Mass education Economic development Division of labour Urbanization A PPENDIX B INVENTORY OF VA RIA BL ES CONSIDERED TO AFFECT THE FERTILITY RATE 43 TABLE Bl VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HCONTRACEPTIVE METHODS, THEIR ACCESSIBILITY AND AVAILABILITYH (OZ) Author Factor Thompson Birth control Carr—Saunders Birth control techniques Coale and Hoover Contraceptive techniques Robinson Fertility control by choice 44 TABLE B2 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY HTRA DITIONAL VALUE SYSTEM PREDISPOSED TO HIGH FERTILITYH Author Notestein Davis Coale and Hoover Petersen (CUSTOM) Factor Traditional home—centred family life Preservation of the family Survival of the group Large family system Family as a productive unit Advantages of the large family Traditional Customs and beliefs Traditional cultural norms Traditional magical norms Traditional religious norms High age at marriage Low proportion of married adults Little remarriage of widows Low frequency of marital intercourse 45 TABLE B3 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY "URBANIZATION" (URBAN) Author Factor Notestein Urbanization Davis Urbanization Blacker Urbanization Coale and Hoover Urbanization 48 TABLE B4 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY "INDUSTRIALIZA TION" (INDUST) Author Factor Carr-Saunders Urban industrialism Notestein Industrialization Factory system Commerce Blacker Industrialization Coale and Hoover Economic development 47 TABLE B5 VARIABLES CONSIDERED TO AFFECT OR SUBSUMED BY "MODERNIZATION" (MOD) Author Factor Carr-Saunders Small family system Notestein Individualism Rising aspirations Educational system Welfare of individual child Protection of individual Davis Individualism Kirk Rising aspirations Desire to limit family size Hatt, Farr, and Weinstein Modernization van Nort Cost of children Petersen Declining average family size Robinson Modern living pattern New value system .3311 71970 vv...‘ [1989