UTAFlTl (New Series) Vol. 3 No.1, 1996:145-160 A Semantic Prototype of the English Word "Responsible" in Tanzanian Usage Stephen T-M. Lukusa. Abstract Though the notion of prototype is better illustrated by natural species and conerete objects, this article argues that abstract concepts can also exemplify it. To prove this, the article investigates the semantic prototype of the English word responsible. In order to evaluate Tanzanian English speakers" judgm~nt of several meaning elements forming this semantic category, the investigation uses a questionnaire based on a set of seven stories. The assumption behind the question concluding every story is that the informants' judgment of responsibility involved can be recorded on a three degree scale. By computing the average score of every story, it is possible to rank the meaning elements composing a semantic prototype for different speech communities, and make relevant comparisons. Introduction To interpret and organize experience, people rely on mental blueprints or categories. By their research in this field, Rosch et al., (1976) revealed that we recognize members of a semantic category by matching them with their prototype or typical example. A semantic prototype gathers a number of properties among which some are central and others peripheral. The semantic category of cars, for instance, can be thought of as vehicles' having the following central features: (i) four-wheeled, and (ii) self-propelled. Therefore, three-wheelers and carts are not typical of cars for two reasons. First, the former are self-propelled vehicles which lack the central property of being four-wheeled. Second, the latter are neither four-wheeled nor self- propelled despite their being vehicles. Though concrete objects (e.g., cars, chairs, cups, etc.) and natural species (e.g., dogs, fish, trees, etc.) are among the best examples of prototypic categories, it is not correct to assume that the concept of prototype cannot apply to such abstract concepts as lie and responsibility. Department of Foreign Languages and Linguistics, University of Dar es Salaam 146 S. T.MLukusa This paper aims at proving this by developing a prototype category of the adjective 'responsible' in Tanzanian English after Coleman and Kay's (1981) model prototype of lie. This prototype category is of course related to those which are found in the other types of English. This resemblance will be better reflected through a comparison between the people's judgment of this English concept in England and in Tanzania. In order to reach this objective, the paper proceeds along the following lines. The first section will search for several prototype elements of the word 'responsible'. The second section will describe the experiment on which the investigation is based. The third section will analyze the results of the experiment. The fourth section will compare the prototype of 'responsible' in the two speech communities. The fifth section will unveil the methodological problems that were met in the investigation, followed by a brief conclusion. 1. Searching for the Prototype Elements of 'Responsible' The first idea that springs to the mind while thinking about the word 'responsible' is that of "guilt" as opposed to "innocence", i. e. , the word "'responsible'" basically means that someone is acc.ountable for the actions (generally bad ones) one is supposed to have performed. This interpretation might not be far from general belief. In attempting to identify the other elements constituting the prototype of this word, I will use my personal experience as well as Hornby's Oxford Advanced Learner's Dictionary of Current English. This somehow random choice was justified by availability of the source when the investigation was being devised rather than by any personal preference. Though it did not bring any new elements, the Lancaster-Oslo-Bergen Corpus (in short LOB Corpus), from which the data attached in the appendix has been extracted helped a lot in checking quickly several uses of this word in context, given the time limitations. The prototype element search revealed the following meaning constituents that were going to be used later in the experiment: (1) (a) X is legally accountable (b) X is morally accountable (c) X knows s/he is accountable (d) X is in a position to be accountable (e) X is to be relied upon (t) X has the obligation to make decisions for Y 148 S.T.M Lukusa ~ 30 • q u • ~ ~ y , 10 d • lIorWla-'- Fig 1: Proportions of the meaning elements in the LOB Corpus data As shown by the adverbs they include, elements (a) and (b) refer respectively to law and morals. One's responsibility is generally determined with regard to laws, and principles of right and wrong . . In element (c), people's belief that nobody is to be blam,ed for unconscious deeds was taken into consideration. An adult, for instance, is better disposed to tolerate a slap in the face from a baby than from a mature person because the latter already knows what slbe has done, and is therefore answerable for ~ - In element (d), one's position sometimes determines hislber degree of responsibility. For example, in a society people feel more committed to helping closer relatives than strangers. Element (e) views 'responsible' from the angle of trustworthiness. Some people, for instance, have such a high sense of duty that others generally trust them because they always do what they are expected to at the right time. In element (t), it turns out that responsibility refers to the obligation to make decisions on behalf of others, e.g., to be responsible for a group implies to bear the charge of deciding its fate. Element (g) has a rather pessimistic view of responsibility. A responsible person appears here to be a scapeg~al. Slbe is blamed for the mistakes of the people of whom slhe is in charge. -- A Semantic Prototype of the fnglish Word "Responsible" 149 It appears from element (h) that 'responsible' is sometimes used to refer to the origin, the source, or the cause of some action-e.g., to be responsible for a change is to have caused it. The above elements of the prototype of 'responsible' are not always so clear-cut as the analysis above tends to suggest. More concretely, saying for example "Who is responsible for this mess in the kitchen?" implies asking who has caused it. This may, as well, amount to looking for who is to be blamed for it. In connection with this, Coleman and Kay (op. dr.: 27) say: Semantic categories frequently have blurry edges and allow degrees of membership. On this view, applicability of a word to a thing is in general not a matter of Yes or No, but rather of more or less. After this analysis the prototype elements of 'responsible', the following section describes an experiment aiming at testing them. 2. The Experiment The experiment supporting this investigation is based on a questionnaire which was pre-tested on a smalr group of eleven native English speakers with a minimum level of B.A. degree. For the sake of comparison, their results will be matched with those of a larger group of one hundred and sixteen Tanzanian English speakers, including indiscriminately undergraduate and postgraduate students at the University of Dar es Salaam. While the British group ranged between sixteen and fifty-five years of age, the Tanzanian one included informants aged between twenty-one and forty-five. The linguistic background of the latter consisted of Swahili, other Bantu languages, and English. The questionnaire was based on seven stories among which one was true, and six fictitious. These stories were carefully selected in such a way as to facilitate a comparison of the prototypes of the word 'responsible' as used in these two speech communities. Each story ended up with a question enquiring about an implicit prototype element. The question sought the informant's judgment of the degree of certainty as regards the prototype element concerned. For the sake of simplicity, the informant's answer was to be given in the form of a three degree scale as shown below. The informant was therefore offered to tick the appropriate answer, thus scoring the grades below. Is X responsible? No Nor roo sure Yes (1) (2) (3) 1SO S.T.M Lulcusa According to Coleman and Kay (1981:30), this scoring system yields a numerical judgment for each informant's rating of each story. The order in which the meaning elements occur in stories does not correspond to that in which they are presented in (1(a)-(h». Since negativity of meaning may have psychological effects on answers, it was preferred to start with a positive meaning of 'responsible' followed by a neutral and a negative one, and to mix all of them so as to avoid any suggestive order. The stories in the questionnaire appeared as follows. I. Mary was a blind person. One day. she fell down and had a bump on the head. Since then she says that she can see and describe any concrete object she couldn't see before. Is thai fall 'responsible' for Mary's recovery? (1) No (2) Not too sure (3) Yes II. GCM is a mining company. It owns a farm which is looked after by Bob and Max. This is the only job these two people have. They are paid by GCM. Is GeM 'responsible' for the management of this property ? (1) No (2) Not too sure (3) Yes III. Once Jack and John had a hot discussion. Jack took a knife from his overcoat pocket and stabbed John to the heart. John died of that wound later. Is Jack 'responsible' for killing John? (1) No (2) Not too sure (3) Yes IV. Bob and Martin are friends. They live in Lonsdale. One night they attended a party. The next morning Bob was so ill that he could neither speak nor communicate by any other means. Is Martin 'responsible' for Bob's health? (I) No (2) Not too sure (3) Yes V. Peter was driving his own car. On his way home from work, he realized that he had left his office key hanging on lhe door and Slopped suddenly. The driver behind him came and hil him inlhe bad. A Semantic Prototype of the English Word "Responsible" 151 If you wer~ P~t~r, would you say that you wer~ 'r~spo1lSibk' for thai accident? (1) No (2) Not too sure (3) Yes VI. Jacob has an account at Barclays. He often gets credits from this bank and never forgets to pay these debts. WoultlYou consid~r Jacob to bt! a 'responsibk' custOlMr? (1) No (2) Not too sure (3) Yes VII. Joe met Pamela reading a newspaper at a local restaurant. They did not know each other before, neither did they have any conversation~ After taking a cup of tea, Joe left Paniela still reading. On the road, Joe realized that be had forgotten his umbrella on the restaurant table. He could not find it when he went back there, though Pamela was still there reading her newspaper. Is Pamela 'r~sponsibl~' fOT that loss? (1) No (2) Not too sure (3) Yes Table 2 shows which meaning elements were present in each story (cf. +). Table 2: Meaning Elements present in each story Stories: I n m IV V VI VB M~aning elements: X is legally accountable: + X is morally accountable: + X knows slhe is accountable: + X is in a position to be accountable: + X is to be relied upon: + X has the obligation to make decision for Y: + X bears the blame for Y's mistakes:. + X is the cause or source of Z: + + 152 S.T.M Lukusa Though each story was meant to test a single meaning element, stories II and III contained two. elements which were not so easy to separate. Let us now turn to the results of the main investigation. 3. Results of the Experiment. This experiment aimed at producing a different ranking of several meaning elements resulting from Tanzanimt speakers' judgment of the degree of responsibility involved in the stories. This supports the viewpoint of cognitive psychology that the same experience of the world! may be interpreted and organized differently depending on mental blueprints called categories. Table 3 presents a summary of the results of the main investigation. Table 3: A Summary Presentation of the Results of the main test ........\:':::~#hiMJl:N~:: :'ik~lt@Aif~r:,¥~::t:j!~b~Hi?:~:: Story I. Mary 38 76 120 234 2.01 n. GeM 30 16 234 280 2.41 III. Jack & John 4 32 288 324 2.79 IV. Bob&Martin 80 48 36 164 1.41 V. Peter 24 32 228 284 2.44 VI. Jacob 12 32 264 308 2.65 VIIJoe & Pamela 72 80 12 164 1.41 Every answer was scored as expl~ined in Table 3. By adding up the scores on each answer, a total score by story was calculated. From this total score a mean scale score was worked out. The mean scale score for each story shows how much the speech community members believed that story to be an instance of responsibility. It should be noticed that the mean scale score was obtained by dividing the total score by the number of questionnaires, i.e. 116. Table 4 compares the ranking of the stories by Tanzanian English speakers to that by native English speakers. A Semantic Prototype of the English Word "Responsible" 153 Table 4: A Comparison in the Ranking of th~ Stories I. m. lack cllobn 2. VI.lacob 3. I. Mary 4. V. Peter S. IV. Bobel Manin 6. D.G.C.M. 7. vn. Joe&Pamela Though this comparison shows that some meaning elements (namely, III & VI) rank similarly in the usage of the two comm~ities. it should be admitted that the majority of elements prove that the hierarchy of meaning elements is highly susceptible to changes depending on .the environment arid people's perception. The histogram in' Figure 2 derived from Table' 4 shows clearly how the investigation led in the two speech communities yields two. different rankings of the stories by comparing the mean score of each. 3.0 2.5 2.0 Jt~ 1.5 1.0 0.5 0.0 ' i J I.Jack&Jo VI.Jacob I.Mal)' V.Peter IV.Bob&M II.G.C.M. VII.Joe&Pa mID score in Brit. Engi IYJJ score in Tanz. Engl Fig. 2: Comparison of the mean scores for the lwo communil ics 154 S.T.M Lumsa Apart from the fact that stories III, VI, and VII keep respectively the same ranks (namely fIrst, second, and last), no story has been evaluated the same 'IIay on both sides. A close examination of why some stories were ranked so high revealed that responsibility in story III was easily recognized by all British English speakers, and a vast majority (82.7%) of Tanzanian informants. This easy recognition is probably due to the number of the prototype elements in this story, Le., apart from legal accountability, it also implied that Jack was the cause or the source of the killing action. In contrast with that, story II was assumed by the author to contain two prototype elements. Yet responsibility was not so easily recognized in it by different English speakers. While the author assumed that GeM's responsibility for the management of its farm implied both the obligation to make managerial decisions and guilt in case of mismanagement, informants on both sides probably recognized the first element only. This lowered story II to rank 4 among Tanzanians, and rank 6 among British English speakers. What does the comparison of the two prototypes reveal? 4. Comparison of the Two Prototypes of "Responsible" Substituting the elements concerned to the stories in Table 4 Yields the comparison in Table 5 of the two prototypes. This comparison shows, first, the order of importance from central to peripheral, followed by the meaning element and the mean scale Score (between brackets) in British English and in Tanzanian English. This agrees with Leech's (1981 ;85) remark that in addition to the category-recognizing ability, human beings also have a different order of cognitive ability-something which is much more closely tied to language-which is the ability to recognize structural relations between categories. It should be made clear that stories III and II which were assumed to include two prototype elements have been considered for one clement for the two following reasons. First, clement (h) was dropped in story II since it did not increase the informants' chances of recognizing responsibility as in story III. Second, element (h) was also dropped in story III because there was an unambiguous (and therefore beller) way of testing it in story I. Hence story III had to be considered as tesling clement (a) which had no better way of being tested. .A Semantic Prototype of the English Word "Responsible" 155 Table 5: Comparative Table of the tw.0 Prototypes ......:.:.:.:...:.:.:.:.:...:~.::::::::::l:;:::I::!!~!i:iii!i:!!!:i!:I::::i:j:ii:::::iii:i::i::!:!:I:i:::iill:: :::~;::II::111111111:1'••1 I'ilill!lrlll'II'IIII'11 Xis legally 3.00 I X is legally 2.79 accountable aCcountable 2 X is to be relied 2.45 2 X is to be relied 2.65 upon upon 3 Xis the 2.18 3 X is morally 2.44 cause/source ofZ accountable 4 Xis morally 1.90 4 X has the obligation 2.41 accountable to make decisions forY 5 X is in a position 1.18 -5 Xis the 2.01 to be accountable cause/source of Z 6 X has the 1.09 6 X knows s/he is 1.41 obligation to make accountable decisions for Y X is in a position to 7 X knows slbe is 1.00 7 1.41 be accountable accountable 5. Methodological Problems The problems encountered in this investigation are of two types: procedural and technical. 1. The procedure adopted in this attempt was not irreproachable. Here are some of the problems that were encountered . • A number of complete questionnaires had to be dropped because they were flIled by non-Tanzanian informants. This waste of 156 S.T.M Lulcusa materials and energy was due to the fact that informants to whom the questionnaire should be given were not selected in advance . • Some questionnaires got lost for two reasons. First, due to my lack of familiarity with the informants, I had to be more diplomatic by persuading even those who said they did not have time to take it to accept to collect it and work it out at home. As a consequence, some informants did not return the questionnaire. Second, the questionnaires distributed by the investigator's assistant were not worked out under strictly controlled conditions. No tracks could be found of some . • Lack of a homogeneous community of native English speakers in Tanzania forced. me to limit the scope of my comparison to the eleven British English speakers of the pre-test . • The reduced number of informants in the pre-test did not enable the researcher to generalize the results with confidence. 2. From a technical point of view, two shortcomings can be mentioned . • The three point answer scale which was adopted for the sake of simplicity did not reflect real psychological distinctions . • The failure to separate some prototype elements (e.g., (a) & (h) in story III and (I) & (g) in story II) did not allow the researcher to evaluate them to their real importance in the prototype. Conclusion We have seen from this investigation that the word "responsible" does not have a unitary meaning. It is rather a complex of meaning elements. The paper has illustrated how the concept of prototype applies not only to natural species and concrete objects, but also to such an abstract notion as 'responsibility'. It has showed how the two speech communities interpret and organize the sl!lDe experience of the world differently. Though the same meaning elements have been recognized as forming the prototype of responsible by the two communities, they have not been appreciated to the same value. Starting from the same central meaning elements (namely, "being legally accountable" and "being relied upon"), either prototype shows a different order of elements. The experience revealed that in Tanzanian English, X is responsible implies that A Semantic Prototype of the English Word "Responsible" 157 I. X is legally accountable 2. X is to be relied upon 3. X is morally accountable 4. X has the obligation to make decisions for Y 5. X is the cause/source of Z 6. X knows s/he is accountable, or X is in a position to be accountable. The results of the experience also proved that in British English, the prototype of 'responsible' has the following order of the same elements. 1. X is legally accountable 2. X is to be relied upon 3. X is the cause/source of Z 4. X is morally accountable 5. X is in a position to be accountable 6. X has the obligation to make decisions for Y 7. X knows s/he is accountable The finding that the hierarchy of meaning elements differs in the usage of the two communities implies that one' s judgment of an experience may not necessarily be shared by a member of a different community. Such differences in people's perception of the world can make communication difficult (if not impossible) accross cultures. Being responsible for the management of a company does not necessarily mean to British citizens what it does within an African context. Bibliography Coleman, L. & P. Kay. 1981. Prototype Semantics: the English Word LIE. lAnguage, 57 (I): 26-44 Hornby, A.S. 1985. Oxford Advanced Learner's Dictionary of Current English. Oxford:OUP. Leech, G. 1981. Semantics. Suffolk:Penguin Books-. Rosch, E. and C.B. Mervis. 1975. Family Resemblances: Studies in lhe Internal Structure of Categories. Cognitive Psychology. 7, 573-605. Rosch, E. el. al.. 1976. Basic Objects in Natural Categories. Cogll;til'e Psychology, 8, 382-439. 158 S.T.M Lukusa APPENDIX The LOB Corpus data The sentence fragments below come from the Lancaster-Os.lo-Berge~ Corpus (LOB corpus in short) which is a collection of spoken and written E~ghsh ~at~ from the British news media. The numbers before them refer to their location In the source, whereas the bracketed letters at the end show the meaning elements they illustrate. A03 197 Ministry or Minister directly 'responsible' for it...(f) A11 169 regard him as 'responsible' for a breach of confidence ...(h) A21 104 solve successfully the 'responsible' tasks facing us. (f) A27 219 They are 'responsible' for a great deal of charity work ...(h) A31 162 Is he not 'responsible' to the whole House?(a) B11 176 ...until a 'responsible' government with an African Prime Minister ... (e) B12 185 On Berlin again the 'responsible' Labour Party view and the Conservative view are so c1ose...(f&d) B 13 110 Wrth his UNO team he has been 'responsible' for more bloodshed (h) B14 172 Executive councils are 'responsible' for the general practitioner, the dentist, the supply drugs. (f&g) B121 185 ...or Mr. Kruschevas 'responsible' for East Berlin ... (f) B25 154 Those 'responsible' obviously have too much time on their hands ... (d&f) 006 112 Even those who held a high and 'responsible' office lived ... (d&f) 0171 149 But it can be reasonably hoped that 'responsible' leaders of those two powerful branches of the Dutch Reformed Church ... (f&g) 017201 ...politics is {politikos, civilis} that which belongs to the citizen as a citizen, and is, therefore, constitutional, and 'responsible' .(f) E28 200 ...water for which the Metropolitan Water Board was 'responsible' for so much contemporary research ... (f&g) E30 134 ...the economic Intelligent unit, which is 'responsible' for so much contemporary research ... (f&h) E37 134 ...milk in two separate six-stall three unit bails with one COwman 'responsible' for each. (f&g) F01 183 ...the doctor who observes signs, and records symptoms and diagnoses the inner states 'responsible' for them (h) A Semantic Prototype of the English Word "Responsible" 159 F02 72 ...an approach to members of the panel 'responsible' for the scheme. (f,g&h) F14 123 Maybe Keith hasn't been as 'responsible' as he should ...(e) F21 7 The belief that a man is as old as he feels is 'responsible' for a great many pulled muscles. (h) F22 151 and 'responsible' for all debts ... (h) F27 12 who would have been individually 'responsible' for resisting the peasants' claims at law ... (a&h) F38 33 ...this fish is probably 'responsible' for innumerable false record ... (g&h) F41 205 There are usually separate buyers 'responsible' for the requirements of each section ... (g&h) F43 77 ...authorities should be active and 'responsible' in the preservation of this country's heritage. (f&g) F44 186 ...is a serious musical student, and is 'responsible' for most of the shadows arranging ... (f&g) G07 162 ...'responsible' for the collapse of the marriage (h) G0898 ...that he was 'responsible' for the strict adherence to recorded precedents ... (h) G24 179 ... was 'responsible' for the poor surface from which Brooklands suffered. (h) G43 43 This means that all 'responsible' should be experts ...(d&f) G43 64 ...he is still 'responsible' for the movements on the stage (which includes arranging that... (f,g&h) G43 72 Moreover, he is 'responsible' for checking the construction ... (f&g) G46 19 ...disorder of her matrimonial affairs, for which she was not altogether 'responsible' . (h) G62 43 he was 'responsible' for selecting ... (f&g) G66 64 sil)ce war in the atomic age no longer presents a 'responsible' and sensible possibility for solving ...(e) G73 165 the principle'of 'responsible' govemment with adult franchise. (c,e&f) G75 174 whether this alleged inferiority of striking power or the conflicts ... is 'responsible' for Kruschev's postponement ...(h) G7672 ... and to set forth conditions of recognition which seem 'responsible' to a watching world. (e) H02 59 ... an independent expert body should be 'responsible' for advising which substance ... (f, g&h) 160 S.T.M Lulcusa H05 96 ...company officials 'responsible' for the management of property ... ~ID . H1074 ...the overseas producer and those 'responsible' for handling and processing ... (f,g,d) H11 29 ...enable Greater London to enjoy an adequate measure of 'responsible' self-government. (e) H1521 ...is not solely 'responsible' for slum clearance ... (h) H25 145 a Christian Child Care Organization should be 'responsible' for their future (f&g) H2925 body which will be 'responsible' for their future ... (f&g) H29 227 any vehicle so left by causing it to be removed in the name of the cost of the student 'responsible'. (h) H30 169 ...too, show that people are living longer. A brighter outlook on the life plus, of course, many far-reaching advances in medicine, is partly 'responsible'. (h) J 13 10 The antibody 'responsible' for the pan-agglutination ... (h) J24 127 ... Projection of shadows, and correlations experiments ... were likely to be responsible for this. (h) J28 29 ... in the event of a breach in the bound all villagers must be equally 'responsible'. (h) J29 97 The school is 'responsible' for behaviour to a great extent... (h) J31 98 their husbands were 'responsible' for the methods ... (h) J31 142 their husbands were 'responsible' for the contraceptive measures ... (h) J39 59 ... How far were the Webbs 'responsible' for their demise (h) J39 97 .. Sidney Webb was 'responsible' for the direction taken by the affairs in 1901 ... (g&h) J40 13 '" the reactionaries created a vast feeling of insecurity, and face with this the 'responsible' authorities generally erred towards the safe side. (d&f) J42 187 However, there are many people, in 'responsible' positions '" (f&g) J6569 the eastern countries were 'responsible' for the manufacture ... (f&h) J75 70 complete explanation ofthe conditions 'responsible' for the similarity ... (h), J77 52 ... who are 'responsible' for the maintenance of ... (f&h)