The Purpose of Field Linguistics PDF

THE PURPOSE OF FIELD LINGUISTICS Field linguistics is primarily a way of obtaining linguistic data and studying linguistic phenomena. 1 It involves two participants: the speaker (or speakers) of a language and the linguistic res...

THE PURPOSE OF FIELD LINGUISTICS Field linguistics is primarily a way of obtaining linguistic data and studying linguistic phenomena. 1 It involves two participants: the speaker (or speakers) of a language and the linguistic researcher. The means of carrying on investigation is the most direct possible, by personal contact. The speaker of the language, the informant, is the source of information and the evaluator of utterances put to him by the investigator. Hence this approach to language study has also been called the informant method and might also be called the contact method (Hockett 1948:119). Field linguistics can be carried on anywhere, not just in the field, as its 1 The term field linguistics is not to be confused with anthropological linguistics which, accord- ing to Carl F. Voegelin (19596, 1961) is coordinate with five other aspects of linguistics, namely, theoretical, psychological, critical, communicational, and comparative. Anthropological linguis- tics "comprises the analysis... of either a wholly unknown language or of some unknown part of a language that is known in other parts" (19596:122). Under this term is stressed the dis- covery part of linguistic research, whether it be of whole linguistic systems (that is, grammars), of certain aspects of these systems (for example, phonemic systems), of differentiation over an area or through various classes of society (for example, dialectology). Hoijer's definition is even more at variance with our concept of field linguistics: anthropological linguistics is that "area of linguistic research which is devoted in the main to studies, synchronic and diachronic, of the languages of peoples who have no writing" (Hoijer 1961:110). 1  2 The Purpose of Field Linguistics name implies. A "field archeologist" must go out to where he expects to collect his data, but a linguist can bring his source to himself. Thus, some field work is done by bringing jungle dwellers to a city and is conducted in an office instead of a lean-to. Field linguistics is generally thought of as work done on languages which have either never been studied before or only poorly, but field work can be done on any language for any purpose. Thus, when a person teaches Vietnamese with the aid of a native speaker of that language—because he himself has less than a native control of the language—and seeks to learn something about the language that the available descriptions give no help on, he is engaged in field work, brief though it may be. But if one wants to restrict the term "field work" to more prolonged or more intensive investi- gations, one can still say that the teacher in our illustration was conducting language research under field conditions. In this book field linguistics will be applied to the more or less intensive study of linguistic behavior. Field linguistics has played an important part in man's study of lan- guage. If we had to depend only on the written records of language, our knowledge would be severely restricted indeed. Fortunately, for several centuries now inquisitive men, some more talented and careful than others, have personally collected data about little known languages and dialects. Not infrequently the investigations were characterized by theories about the origin and nature of language and the relationships between languages. As inquires about language became more and more sophisticated, there was an increase in the direct study of living languages. Oftentimes this was of dialects of the cultural languages or of their culturally less important "sister" languages. And although it was a long time before some language scholars could look upon the "primitive languages," as they were known, with any degree of seriousness, there was nonetheless a considerable amount of research in non-Western languages. Some of this was quite good even by our present standards. But since the beginning of this cen- tury the amount of field work engaged in has risen dramatically. It is not coincidental, moreover, that this same period saw an equally dramatic development in the science of language. In fact, the history of linguistics cannot be told without proper recognition being given to the contribution made by linguists working on living and for the most part non-Western languages. The coming of age of linguistics has in no way diminished the impor- tance of linguistic field work. If anything, the need is even greater, because we see its value more clearly. There is today no need to justify linguistic field work in general. There are literally hundreds of individuals currently engaged in some form of field linguistics. The money being expended in this research amounts to several millions of dollars each year. By way of 3 The Purpose of Field Linguistics illustration, however, there are four ways in which field linguistics can serve some important purpose. 1. There is still a dearth of basic information about the languages of the world. We are not even certain exactly how many languages there are. The more we learn, the greater the number becomes. About thirty years ago a very rough estimate put the number between 2500 and 3500 (Gray 1939:418), but a more recent guess was that it was between 4000 and 7000 (Ferguson 1964). In West Africa alone, where some people thought there were 300 languages, field work led one investigator to raise the number to "well over 500" for 15 countries south of the Sahara from Senegal to the Cameroun, and there certainly are areas more linguistically diverse than this one (Ladefoged 1964:xiii). There are, of course, languages all over the world which are disappearing with the passing of their monolingual speakers or with the assimilation of their speakers into a dominant society through bilingualism. For example, of the 181 North American Indian languages in use today, 49 have fewer than 10 speakers of whom most are over 50 years old (Gursky 1963; see also Chafe 1962, 1965). We shall be the poorer for not having studied the dying languages. As for linguistic descriptions of one type or another, most languages are hardly known. Even the 11 languages which together account for over one half of the world's population are inadequately described (Ferguson 1964). The other languages suffer more acutely. It is too much to expect all languages of the world, or even all languages spoken by one million or more people, to be fully described, but descriptions of specific aspects of grammar (for example, sound systems, grammatical categories, clause types) for many languages are realizable. The gathering of all this informa- tion is incontestably the business of linguists. 2. Field work is indispensable for the development of linguistics. As put by F. G. Lounsbury (in C. F. Voegelin 1950:299), "The recording and description of every one of these [dying languages] is as important to the science of linguistics as is natural history to the science of biology. When the 'natural history' of language is fairly adequately written,... we can look forward to much more of a 'science of general principles' in lan- guage than we now possess." The following paragraphs reveal only a few ways in which field work can lead to the development of linguistics. Linguistic research can contribute data toward the understanding of > language universals. There have been recent attempts to delineate some of these universal characteristics of language (Greenberg 1963), and field work since these publications has already challenged some of the assertions (Ladefoged 1964). When languages have been described and the informa- tion about them is easily retrieved, there will be less excuse than there is today for statements such as "There are many languages where... , " , "There may be languages in which... ," and "The possibility that some  4 The Purpose of Field Linguistics languages do not clearly distinguish..." Thus, until K. L. Pike had done field work on Campa, an Arawak language of Peru, it had been held, even by him, that there were no languages in which contiguous syllables could have primary stress, but Campa has series of three and occasionally of even four primary stresses, such as ——-—-—' (Pike and Kindberg 1956). Even more significant is recent concern with the identification and use of language universals in the stratificational and generative-transfor- mational models of language (Teeter 1966). There is also a direct relation between field work and language descrip- tion. The more field work, the more information we will have about the variability of language. This information will reduce the amount of time " required for arriving at the distinctive structures of languages being de- scribed. The Campa case serves to illustrate this point. It presented special difficulties to the investigators because they "were not psychologically prepared to recognize the nature of the system when the evidence began to appear" (Pike and Kindberg 1956:415). Progress is spiral and cumula- tive. As put by John Lyons (1962:127):... the history of science is full of examples to support the opinion that the actual cannot be properly described, perhaps not even recognized, except in the framework of what has previously been envisaged as possible. At the same time, of course, the sphere of what is thought of as possible is being constantly revised under the impact of discoveries made in the description of actual languages. The field is also the laboratory of the linguist. First, because in his at- tempt to write a grammar which will be comprehensive, he must test every generalization. This is done by checking statements with informants and with texts. The texts can be studied away from the field, but their validity as a measuring rod is itself a matter to be decided after some field analysis. This means that without proper precautions, a linguistic generalization may be valid only for a specifiable corpus and not for the language as a whole. Secondly, the field serves as a laboratory for all of linguistics insofar as all theories and generalizations are tested by new data. Without these data we will not easily discover the Achilles' heel of whatever theory may be in vogue. Field work is also an antidote for excessive theorizing. Theorizing be- comes excessive when the same problems or the same data are looked at again and again at the expense of ignoring other significant issues. 3. Other disciplines besides linguistics depend on linguistic field work for data, experimentation, and problem formulation. Anthropology is one discipline which early recognized its dependence on it. The two sciences— anthropology and linguistics—in fact developed in the United States each supporting the other. The importance of language to ethnographic field 5 The Purpose of Field Linguistics work is first of all that of a tool. As put by Boas (1911:60): "A command ; of the language is an indispensable means of obtaining accurate and thorough knowledge." (Other ethnographic field workers, however, have argued among themselves about the skill that they ought to demonstrate in the use of the "tool language," some pushing for fluency and others for a minimum control necessary in interviewing. 2 ) For a long time the collection of texts in the native language constituted a considerable part of ethnographic field work. The text publications of Franz Boas, for example, run up into thousands of pages (White 1963). Ethnographic texts are an important source of explicit information about the culture being studied (Radin 1949). They also serve as a stimulus for particular investigations, for in the process of translating the texts the ethnographer sometimes discovers valuable clues. Thus, Suttles discovered a set of kinship forms in his texts which could not have been elicited by the genealogical method, because they cannot be used with possessives (1965 : 161; see also Lounsbury 1954:226). Radin (1949) points out anothervalue of text collection. By being able to write down and then read the text with a fairly good approximation to the informant's pronunciation, the ethno- grapher gives the impression of understanding what he has obtained. En- couraged by this comprehension, the informant, who perhaps has hitherto been psychologically at a great distance from the investigator (or the race or class represented by him), becomes more interested in giving him information. Anthropologists are concerned today with many problems—historical, structural, and functional—which must be studied linguistically. What, for example, are the correlations between various aspects of cultural be- havior and semantic structure; the relation of perception to linguistic structure; the semantic structures typical of specific cultures or universal in all cultures? For each problem there may be necessary a specific set of data obtainable only from native speakers. For such work investigators must be sophisticated in both anthropology and linguistics. But anthropology is not the only science to profit from the techniques of linguistics. Bloomfield once said (1925:1): "The science of language... is most closely related to ethnology, but precedes ethnology and all other human sciences in the order of growing complexity, for linguistics stands at their foot, immediately after psychology, the connecting link between the natural sciences and the human." All that has happened in the intellec- tual world since then has only confirmed his assertion. Today linguistics cannot be ignored by philosophers, logicians, psychologists, or theolo- 2 For example: Mead 1939; Lowie 1940; Beals 1957; P. Bohannan 1958 a, b; McEwen 1959. For a statement as strong as Boas' see Radin 1949.  6 The Purpose of Field Linguistics gians. Some of the questions they pose to linguists can be answered on the basis of what we already know about language and its use. Others must await further research. 4. Finally, field work is necessary if linguistics is going to be applied practically to human affairs. For example, with the eclipse of the colonial era and the rise of new nations, there is an urgent need for language plan- ning and "language engineering." Nations need to determine how many languages are spoken within their boundaries, which ones should be used for education and other purposes, and how they are to be adapted to modern life (Le Page 1964). Some of these languages, like Arabic, have an ancient literature but are fragmented by dialects. Others, like recently resuscitated Hebrew, must be fully equipped for the scientific era. Trade languages, like Malay, Fula, and Sango, must be standardized. There are also vernacular languages which only recently have been raised to official status. For example, the declaration of Pashto as an official language of Afghanistan along with Persian in 1935 (Shafeev 1964) meant the estab- lishment of an Afghan Academy for the purpose of carrying out research on the Pashto language, folklore, and literature. The academy has also sponsored the publication of Pashto classics, folklore collections, gram- mars, and dictionaries. In this century we shall undoubtedly see a whole series of linguistic revolutions, some the effects of political developments but others the re- sult of the population explosion. If "minor" languages were so classed because of the number of their speakers, some of them are becoming less "minor" because of the population explosion. They will demand more and more attention from linguists or linguistically trained people.  HUMAN FACTORS IN FIELD WORK In his preoccupation with data—in the tangible form of notebooks, slips of paper, lists of words and sounds, and so on—the linguist can very easily forget the human factors in his investigation of language. His collec- tion and analysis of language phenomena are dependent on and in some way influenced by the people among whom he works and by his own per- sonality and training. Without some understanding of himself, the lan- guage community, and the informant the linguistic investigator goes ill-prepared to the field. So much has to be said about the informant that the whole of Chapter 3 is devoted to this subject. The Investigator Not everyone who is attracted by the study of language is qualified to carry on linguistic research in the field. Veteran field workers can tell of investigators who have increased the difficulties of their endeavors by their insensitivity to the subtle elements of harmonious interpersonal relations. Anticipating problems, many have fulfilled their linguistic ambitions in studies where human factors were less crucial. This is not to say, of course, 7  8 Human Factors in Field Work that outgoing persons cannot be found among those whose studies are less dependent on the first-hand collection of data from speakers of a language. It is simply a fact that one who is embarrassed by too much exposure to the challenges of interpersonal and cross-cultural relations can nonetheless find satisfying pursuits in other fields of linguistics. To the field situation the investigator brings not only his personal psy- chological makeup but also his training and skills.1 There must be training in articulatory phonetics and methods in linguistic analysis—phonologi- cal, grammatical, and to some extent lexicographical (or semantic). Occa- sionally one finds statements by linguists who in describing their field techniques seem to undercut the need for intensive training in phonetic transcription. The following is one example. It is said (Voegelin 1949:81) that whether the investigator starts with a phonetically over-differentiated [that is, with narrow transcription] or under-differentiated [even to the point of leaving out contrastive sounds] text, he must "check again and again with native speakers before arriving at a text identified entirely in terms of phonemes." If this is a procedure to follow in field work, it is (a) extremely wasteful of the time of both the investigator and informant, (b) unnecessary, and (c) perhaps even detrimental to the research, since it would tend to undermine the informant's confidence in the investigator. Checking will always be necessary, but it should never be done on a text in such an unsystematic way. The more that one knows of the diversity of language structures and of the different ways they have been handled by grammarians, the more successful he will be in dealing with the language at hand. Reading of various types of linguistic descriptions should therefore be an integral part of linguistic training. Unfortunately, finding good grammars is not ac- complished by an examination of the reviews. Rarely do grammars re- ceive accolades like the one Hockett gave Bloomfield's Algonquian work: "one of the finest indoctrinations into the best of linguistic method" (1948: 117). Linguistic "internship" is practice in the analysis of a language. Know- ing the principles of linguistic analysis—as demonstrated by a lecturer, revealed in the reading of descriptions, or worked out in problem-solving —has proven to be inadequate for all but the rare individual. 1 There will, of course, always be gifted people who can collect data and write a description which is accurate and comprehensive without the benefit of formal training in linguistics. One such person was Lt. Col. D. L. R. Lorimer. This "amateur linguist," whose major responsibili- ties were those of an administrator in the British civil service, wrote a grammar of Burushaski of which it is said: it is "one of the best 'amateur' efforts that has ever appeared in linguistics, and is marked by acuteness both of recording and of analysis." Lorimer, it is claimed, "had little to learn... [in field methods] from the professionals, apart from the field of phonetics and some of the very latest developments in phonemic and morphophonemic analysis" (Emeneau 1940:354. The Investigator 9 If intensive training in field procedures is assumed for the anthropolo- gist, it can figure no less in the training of a linguistic researcher. Yet in some quarters there seems to be a minimizing of its importance. In a paper read to a conference of anthropologists, linguistic analysis was made to seem an easy task for them: In order to do his linguistic work, accordingly, the anthropologist does not need to be trained in a whole separate science. He has only to comprehend the phonemic method of discovering what sounds are functional in a given lan- guage [that is, what are the phonemic contrasts], and the combinatorial tech- nique of stating which combinations of linguistic elements occur in speech" (Voegelin 1952:325). 2 What this statement says is that a skilled ethnographic interviewer who is acquainted with the principles of linguistic analysis should have no diffi- culty in producing a satisfactory grammatical description. The implication is that ethnographic and linguistic techniques are fundamentally the same; one must only add some supplementary skills to those he already has. It is doubtful, however, that the results will always be the desired ones. For the anthropologist as well as the nonanthropologist there must be training, and it will be incomplete without practice in linguistic analysis, using an informant. The field worker's preparation must also include becoming acquainted with the culture of the people whose language he is going to study and with previous work on this language or others either related or found in the same area. This is the testimony of one linguist in this regard: "I once attempted a grammar... of a single idiolect of Kalmyk Mongolian at a time when I knew little of present-day dialects. The result of the analysis seemed linguistically adequate; it was only after studying related idiolects and dialects that a full awareness dawned on me of the vital importance in linguistic field work of some background in—and feeling for—the family of languages involved. (It is appalling how many hours one can waste working on problems already solved by others.)" (Street 1963:336). The reason that ethnographic reading is important is that in language analysis one must make inquiries that will be judged culturally relevant or irrelevant by the informant. Irrelevant inquiries will have the effect of slowing down the elicitation and perhaps of undermining the informant's confidence in the researcher. When eliciting words from the pastoral Tamachek of northern Mali, for example, one does not simply ask for the generic terms for domesticated animals: camel, horse, cow, and others. The investigator should expect that the Tamachek would have a profusion 2 This statement appears to contradict an earlier one: "Everyone acknowledges this [that is, that language is part of culture] theoretically and then tends to treat the two separately in actual work because the techniques of gathering data and making analyses are not the same for both" (Voegelin and Harris 1945:456-457).  10 Human Factors in Field Work of terms in this lexical domain distinguishing male from female animals, young from mature, those in good condition from those in poor health, and so on. The more relevant to the culture that a linguist can make his inquiries, the more interested his informant becomes. The Community Linguistic field work does not have to be done in the native habitat of the speakers. Many Americans have worked with single informants who were away from home. Thus Morris Swadesh (1951Z>:66) was able to elicit about 500 words, a few paradigms, and a short text from a Unaaliq Eskimo at a Sportsmen's Show in New Haven, Connecticut, in 1936. Idiolectal grammars have been the target of repeated criticism, but there is no denying their contribution to our knowledge of the languages of the world. Nevertheless, if linguistics is going to concern itself with more than structural descriptions, as indeed it should, the linguistic researcher must expose himself to the normal use of the language. There are two principal reasons why field work should be done in situ, not necessarily with several different informants (whose number is taken up in Chapter 3), but where one can see people realistically interacting in the language being studied. The first reason is that field research will make it much easier for both the informant and the investigator to collect a corpus which is culturally relevant and linguistically accurate. A careful study of examples in grammars often reveals utterances which quite clearly show the bias of the investigator. Examples of poor analysis are more difficult to detect unless the reader happens to know the language as well as or better than the analyst. The second reason is that by restricting himself to working away from the community, the investigator precludes the observation of important linguistic phenomena. It is not just that the "meaning" of utterances can be ascertained with greater accuracy—and meaning discrimination is an important tool in linguistic analysis—but also that there is much data of sociological or psychological significance that can be obtained in no other way. It is becoming increasingly important for the linguist to be able to relate his findings to other fields of linguistics and to other disciplines. An example of the limitations of the informant technique is my own linguistic investigation of Gbeya dog names (1965). My purpose in collect- ing them, along with personal names, proverbs, and riddles, was primarily to increase my vocabulary and to investigate the possibility of there being different styles of language. Except for a few syntactic features, my in- vestigation turned up nothing linguistically significant, but many new in- sights into Gbeya culture were obtained. When these names were being The Community 11 described for publication, however, I discovered that I had very little information about their most interesting aspect from a theoretical point of view. What turned out to be extremely interesting was not their lexical content or syntactic structure but their use in bringing attention to the behavior of others or to one's own attitude to life. Entrance into a community can be extremely difficult. In a few instances investigators have lost their lives even after great effort was made to establish rapport with the people. More common than danger to life or limb is the problem of being ac- cepted by a community. Communities vary considerably in their attitudes toward outsiders. Some are extremely hospitable and others treat strangers with great suspicion and lack of cooperation. Some, like the Auca Indians of Peru, consider the approach of any person outside the immediate homestead a threat to life. More typical is the treatment afforded strangers by the people of Sirkanda, a peasant village of North India. The following is a statement by an Indian school teacher from fifty miles away who was assigned to Sirkanda: I have taught in several schools in the valley and people have always been friendly to me. They have invited me to their homes for meals, have sent gifts of grain and vegetables with their children, and have tried to make me feel at home. I have been here four months now with almost no social contact aside from my students. N o one has asked me to eat with him; no one has sent m e so much as a grain of millet; no one has asked me to sit and talk with him; no one has even asked me who I am or whether I have a family. They ignore m e (Berreman 1962:5). Another fairly well described case of cautiousness, suspicion, reticence, and sensitiveness is that of the Gullah Negroes who live along the coast of South Carolina and Georgia. Their language had never been adequately studied until Lorenzo Dow Turner, himself an American Negro, estab- lished rapport with them in the forties. His work Africanisms in the Gullah Dialect has been acclaimed for its demonstration of what good field techniques can accomplish. Another reason for a community's reticence in accepting a stranger is its inability to understand the nature and purpose of a linguistic project. People whose days are taken up with caring for their own needs do not understand the presence of a foreigner who comes to simply ask questions. Even in our own society where interviewing as a basic technique in re- search is known superficially by a large proportion of our society, one's avowed research goals are invariably questioned. When, for example, I was using the sentence-completion device to get information in the United States about English syntax and had quite frankly explained what I was doing, I was still asked: "Is this some kind of a trick?" One might expect people to be flattered by having their language stud-  12 Human Factors in Field Work ied, but this is not always true. Lack of interest is in fact to be expected where one is studying a nonstandard dialect or a language which does not have the prestige of the national or official language. People may fear revealing a bumpkin mentality or ungrammatical speech. Since it is with language that people transmit information about their culture, they may also resist its study for fear of revealing secrets about themselves. Un- sophisticated users of language do not see the difference between structure and content, so questions about their language are taken to be questions about their lives. It is not always obvious to a linguist how he gets into the back region of the community as he investigates its language. His work is certainly not comparable to that of the ethnographer who must get bits of information that have cultural relevance by direct means. But the linguistic investigator also handles bits of information which can take on special significance when he is working in the community. He might innocently ask his in- formant to translate a sentence he overheard. Embarrassment or even anger would indicate that the informant considered this repetition equiva- lent to tattling or slandering. The linguist must therefore expect to arouse his informant's emotions by calling his attention to the content of utter- ances. This is also true as one works with traditional utterances such as proverbs and tales. One of the times my own Gbeya informant was most severely criticized for helping me to learn Gbeya was after I had publicly used the proverb hlrkx] à deà b'ârâ wl-ré ne rékét â deà wl-zu "Escaping makes a righteous person but getting caught makes a thief." The com- munity's resentment must have been based on being "defeated" in an argument with one of its own weapons, a proverb. Illustrative of a community's language shame is the case of the Coatlân Zapotec of Oaxaca, Mexico. Field work among these people is described in the following words: From the initial language contact with the Coatlân Zapotecs it was evident that the language, as far as the people were concerned, branded them as "indios," as ignorant peasants. Progress up and out of this social level was to come through the acquisition of Spanish. A village school, "primaria," had been in operation for some twenty years. Hence for a foreigner to enter the village and express interest in their native language, to learn to speak... , to write it with an alphabet that looks like Spanish—all this for the people was incomprehensible. In fact, the attitude towards our linguistic efforts became that of sustained aloofness and, for a few, hostility. To recognize the Zapotec language as a language and worthy of being studied was interpreted as an effort to regress to a previous generation—the time, in the memory of not a few adults, when the presence of the "mestizo" in the village would find every door shut, with the women and children inside, and only a few of the bravest men outside to exercise their meager vocabulary of Spanish (Robinson 1963:iii). The Mazatecs of Mexico also hold their own language in contempt. The Community 13 Eunice V. Pike reports (1956:73) that the word for "Indian language," meaning their own, is a compound consisting of the word for "word" plus an adjective which is also used in describing a cripple, an idiot, and a person lacking in money or goods. Finally, entry into a community will be affected by its power structure. One must expect some kinds of "political" forces whether the community be formally structured or not. They may be kinship or occupational groups, political parties, religious castes or leaders, and others. In a less overtly structured society, such as one finds in many parts of Africa, one does not necessarily find less opposition. Cliques and old rivalries may stand in the way of any productive work whatsoever. In the novelized description of ethnographic work among the Tiv, Laura Bohannan (writ- ing under the pseudonym of Elenore Smith Bowen, 1964) very effectively portrays the frustrations experienced by investigators as they become involved, sometimes unwittingly, in communal problems. Acceptance by a community even under the best of circumstances is not achieved without some effort on the part of the investigator. He must reveal himself as a harmless individual who in no way threatens the exist- ing state of affairs. This means fitting into the scheme of things as a neutral person or as one who can actually contribute something to the com- munity's well-being. To accomplish this the investigator must assume a role which is acceptable to the community. Since acceptability is deter- mined by the community's expectations, it may take several weeks or even months of rather intensive and systematic investigation to ascertain what these are. Of his work at Sirkanda, Berreman says (1962:9): "Three months were spent almost exclusively in building rapport, in establishing ourselves as trustworthy, harmless, sympathetic, and interested observers of village life." But the field worker, whose time is always severely re- stricted, will want to use every opportunity to the greatest advantage. When this determination affects a person's behavior, making out of him an efficient, and perhaps impatient, machine, he will violate the ideal behavior demanded by the community. The people of Larteh, Ghana, for example, required one investigator "to be... pliant, passive, easy-going, docile, a follower rather than a leader" (Brokensha 1963:533). The variety of experiences people have had in finding their place in a community is truly astounding. Wayne Snell and his family had to be adopted by one of the Machiguenga families since the only other alterna- tive among these Peruvian jungle dwellers was to be considered and treated as an outsider. Benjamin Paul and his wife were given a role by the Guatemalan Indians on the day of their arrival, for the people thought that because of the amount of goods they had arrived with they surely must be merchants! (Paul 1953:433). The case of the Snells in being adopted into a family is an extreme form  14 Human Factors in Field Work of an otherwise effective way of finding an entrance into a community, that is, sponsorship. This simply involves being endorsed by someone in the community. The sponsor does not necessarily have to be a promi- nent or influential person; such a person sometimes creates more prob- lems than he is worth. On the other hand, one must not be sponsored by a person against whom there is resentment. Acceptance by a community can also be achieved by various avenues of participation: (a) The first has already been mentioned—membership or relationship of one type or another—in a family, club, society, work gang, ritual team. Some of these are more or less permanent, but others may be temporary. One recent investigator of the Yupik (Eskimo) lan- guage on St. Lawrence Island, Alaska, became a regular member of a boat hunting crew at the invitation of its captain. Through two hunting sea- sons, four or five days a month, he had the full responsibilities of a crew member. "... 1 was expected to be ready to go out with the crew on a moment's notice and take part in all the work involved on a hunt. This included getting the equipment down to the beach, preparing the boat for launching, implicitly following the orders of the boat captain, doing my share of the butchering, and finally helping land the boat and securing it. All this entitled me to have an equal share in the game that was taken on the hunt." In this natural setting the field worker escaped the stilted conversations which had characterized all his previous contact with the Eskimos, and he was able to note linguistic material which had been dif- ficult to elicit or which had not even occurred to him, such as various types of humor. (He also discovered that the Eskimos continued the practice of casting certain parts of animals back into the sea as an offering to the goddess who releases animals for the hunters to capture.) (b) The avenue available to everyone is the solidarity he has because of his age or sex with certain members of the community. A man, simply because he is a man, has the right to sit in the men's house among the Damal of New Guinea. There are very often certain obligations, of course. Among some people it may mean bringing tobacco. Among the Damal it means being properly and modestly dressed. To the Damal men, indecent ex- posure, comparable to the exposure of the genitals among us, is an un- covered head of hair, (c) One can also participate in some communal activity appropriate to one's sex like drawing water, planting rice, setting traps, and the like, (d) Finally, one can render services which the people would otherwise be without, such as pulling teeth, writing letters, and acting as intermediary with government representatives. Participation exacts a great price from the held worker. There will be physical discomforts as one lives at a level and under circumstances to which he is unaccustomed. More trying will be the obligation to behave in an unaccustomed way. Everyone wants to indulge his own idiosyncra- The Community 15 cies from time to time. In our own societies we know which ones we can favor, where and when. In a foreign community it may be difficult or impossible to "do what we want." The toll that participation exacts of one's time is another factor that one must face. Getting alone to do one's work, as many Europeans are accustomed to, may be difficult indeed. Privacy in many communal societies is unknown as we think of it. Sending people away would be taken as an affront. My own solution to this prob- lem did not come until I had been in Africa for several years. If my visitors wanted only to chat at a time when I was otherwise preoccupied, I told them that I—speaking about myself in the third person—had gone to weed the garden. By using a common phrase, I conveyed the message that although I did not have a real garden to weed, I did have work which kept me equally busy. The good will with which this explanation was always received testified to its cultural relevance and effectiveness.3 The greatest danger in participation is that it undermines the psycho- logical distance that is necessary in scientific investigation. What is studied must be "out there" where one can look at it objectively and holistically. The linguistic field worker must never become so involved with people that he loses sight of the object of his study. Even as he talks with people in the target language, he consciously filters out, in addition to the mes- sage, all kinds of information about sounds, use of certain affixes, phrase structure, and gestures. At some point or another the investigator will have to explain what he is doing and why. In other words, he must identify his role in the com- munity. Inevitably, this role has a direct bearing on (1) to whom he has access or (2) from whom he is excluded and (3) on what he is or is not told about persons and activities. If he has the backing of some local govern- mental body, his project will not be difficult to explain, but where govern- ment officials and government interference are looked upon with distaste and distrust the investigator might be advised to omit references to the government support. Another danger of sponsorship by an external body is that the investigator's work may be misrepresented. For example, when Hollingshead did his study of "Elmtown" youth, he had been preceded by the Chicago Committee on Human Development which had informed the leading people of the town that the study to be undertaken was pri- marily interested in the development of character in boys and girls. By "character" the people understood "goodness." For this reason they were very much in favor of the projected study, but Hollingshead was pri- marily interested in the sociological correlations of the behavior of youth, 3 This paragraph has mentioned a few of the many cultural differences that make it difficult to establish and maintain rapport with members of another society. Shelton (1964) discusses one that is easily overlooked, that is, inability to make physical contact with village folk.  16 Human Factors in Field Work not in prescribing good behavior. The result of this misinterpretation was a limitation of what he could do (Hollingshead 1949:8). If the in- vestigator declares that he is alone in his project, his words may be taken with incredulity. What the investigator must look for is some indepen- dence mixed with endorsement from a well-received prestigious institu- tion, be it political, educational, or commercial. Language study for its own sake will hardly ever be comprehended. More acceptable are explanations in terms of what the study could con- tribute to education, mass information media, and the learning of the standard language. When I was conducting a sociolinguistic study on the use of Sango in the Central African Republic, I explained that although French was the national language, not everybody knew French. There- fore it would be a very good thing for the government and others to as- certain how many people knew Sango. This was one thing I was trying to do. The role of the "learner" is a very effective one. Among the Japanese, for example, a researcher who presents himself as a student of human behavior will probably find, as one sociologist did, that he is equated with a "scholar," a very respected person in their society (Wax 1960). The role of learner has the advantage also of increasing the intimacy between the investigator and his informants. However, in preliterate societies, where there is no scholarly tradition of any type, the image of a learner is impossible except in a personal, practical sense. In such societies the investigator is obliged to give the appearance of being actively engaged in worthwhile projects, especially if the community has a low opinion of "dawdlers." This might mean engaging in red-herring studies, that is, doing valid work which is approved by the community but which is not central to the project. There are obviously questions of ethics when one assumes a role and states a purpose in a community. The first one concerns the ethics of role-playing itself. Is it deceitful to assume a role which is in conformity to the local role expectations even when these are far removed from the explicit purpose of the research? The answer to this question will be de- cided in part by the amount of disparity there is between the role and the purpose. It will be decided also by the expectations of the community. It is not unethical to act in harmony with these expectations. A more specific question is this: how can a linguistic investigator take from the people of a community a vast amount of data—much of it given free, all of it given in good will, with the hope, perhaps, that it will do them or their children some good—and use it exclusively in scientific publications which in themselves can serve no practical purpose? To some degree the linguist is obliged to his helpers to meet their expecta- tions. Looking upon linguistics as an "objective" science does not make us The Community 17 less dependent on human beings for its pursuit, nor does it make us less obligated to use our findings for the satisfaction of their desires. How all of this is done is, of course, a matter of personal decision. One linguist, a rare one indeed, begins his notes to the user of his dictionary with the following words: "Many times during the course of fieldwork the author's Klamath friends and informants expressed the hope that the Grammar and Dictionary might be of use to them as a guide for writing Klamath and also as an instruction book for younger Klamaths wishing to learn the language." Then follows an apologetic statement concerning the pur- pose and use of the dictionary which is fully in good taste (Barker 1963: 10). With the greater availability of education and literature in the newer nations scientific field workers, especially ethnographic ones, are coming into greater and greater disfavor. One African spoke sarcastically about how he had been invited to tea whereas the real purpose was to get in- formation from him. It did not seem that he criticized the elicitation of information nor how it was used; he seemed rather to look upon his con- versation as a commodity over which he should have had some control. "When people make a study," he added, "they should leave behind a copy of all that they write down." Another African discouraged a mis- sionary from studying anthropology after he had himself taken such courses. Similar attitudes are held concerning linguistic work. Two non- Africans saw the wisdom in criticizing, by implication at least, the work that many had done on African languages. At a conference in Africa at which Africans were present John Spencer declared that African lan- guages have been looked upon by academics as "rich linguistic raw ma- terial awaiting export for processing" (1963:37). At this same conference Pierre Alexandre said (in my translation from the French): "Some [Africans]... experience suspicion towards non-Africans who are inter- ested in African cultures and languages, seeing in this interest a disguised means of keeping Africa at the level of the folkloristic and picturesque so as to exploit it" (Spencer 1963:57). In a politically sensitive Africa there are frequent reminders to the field worker to be circumspect about the role that he assumes and the use to which he puts his study, but in other areas there are far fewer demands for alertness. Thus, investigators among the recently pacified groups of Brazil or New Guinea will find it difficult to think of their work in a his- torical perspective. Yet resentment against academic exploitation or "cultural imperialism," as it is being called today in some areas, can be retroactive, which is all too evident in the newer nations. The fact that two well-known Africans were at one time linguistic informants, namely Kwame Nkrumah and Jomo Kenyatta, should prevent field linguists from being too casual about their obligations.  18 Human Factors in Field Work Gross deviations from our society's ethical code, of course, we do not expect. It is therefore surprising when we read in an anthropology text- book: "He [the house-servant acting also as informant] was not always scrupulous, but he got me what I wanted and needed. His methods and motives perhaps would not stand up under close inspection, nor mine either. But moralizing seems inconsistent with the hard facts we had to deal with in the bush." One would hope that no amount of casuistry will ever be adequate to justify the devious activities implied in that confes- sion. Away from the field, that is, in publication, a certain standard of uprightness is assured by a large and critical audience. In the field there are no such sanctions to control the behavior of the investigator. Each man is his own king, but he must also be his own judge. 4 The reader has by now come to the conclusion that field research can be a difficult experience. It can indeed. The reason is that the researcher must cope with innumerable problems, some of which he has never before had to face and others he is very much unaware of. It will be an entirely new and trying experience, for example, to be completely responsible for one's own water supply, lighting, and sewage disposal. Even more difficult perhaps will be the problems which arise in his involvement with people of a different culture. The point is tellingly made in the an- thropological novel Return to Laughter by depicting the field worker's trials during a smallpox epidemic: It is an error to assume that to know is to understand and that to understand is to like. The greater the extent to which one has lived and participated in a genuinely foreign culture and understood it, the greater the extent to which one realizes that one could not, without violence to one's personal integrity, be of it.... I had discovered that there were moral values which I could not willfully abandon, no matter what the dictates and interests of science and no matter how impossible it had been for me to live up to those standards (Bowen 1964:291, 289). The problem which the field worker least expects, therefore, is a psy- chological one. For many a neophyte researcher field work will resemble a birth experience his student life has poorly prepared him for. Ejected from the university's womb into a foreign culture where the most im- portant values are in personal relations, where technical or scholarly com- petence is either not understood or not appreciated, the field worker will indeed feel out of place. He will find himself asking questions he had never heard raised: What is the relation of my project to humanity's cru- cial needs? By pursuing this project am I most adequately fulfilling my own personality and destiny? Am I truly qualified to interpret these people 4 For recent discussions of ethics in field work see Barnes 1963, Hanna 1965, Shelton 1965. Published discussions by linguists have not come to my attention. The Community 19 with my inexperience? Why do the people see me more and less than I truly am? For one sociological field worker the most difficult aspect of field re- search was the discovery of a role which was satisfying, in being authentic, both to herself and the people she was studying. In assessing her own difficulties twelve years after the experience she writes: "... the funda- mental process involved in the initiation into full-scale field experience does not differ significantly from the process of learning any new role. The neophyte passes through an initial period of anxiety and distress, during which he and the persons with whom he interacts contribute to the definition of both his and their roles. If both parties are able to agree on these definitions and are able to communicate the fact that they agree, much of the anxiety disappears and is replaced by the complex phenomena which we call 'self-confidence'" (Wax 1960:177). Her problems were due to the initial failure to identify this role: "... during this period of initia- tion she [writing of herself] found herself trying to defend what to her was an ideal rather than a real self-image against the community's already well-defined derogatory appraisals. Desperately she tried to see herself as a 'person to whom one gave information,' while everyone about her regarded her as 'a person to whom one gave no information'" (175). Her frustrations became grave. Very frankly she describes them with these words: "She [the author herself] spent days alternately crying or writing letters to relatives and academic friends. Then she refused to associate with the Caucasian staff members, hoping that this irrational rudeness would make her more acceptable to the Japanese. She repulsed the over- tures of a few kindly and insightful staff members who tried to draw her out of her self-imposed solitary confinement, interpreting their offers of friendship as 'attempts to steal her data.'... Finally, she succumbed to an urge to eat enormously and in three months gained thirty pounds" (175). The first way to avoid some problems and to cope with others is to be fully aware of them. One must understand the nature of the psychological crises one will have to face. The second tactic is simply to avoid the or- deal of having to work alone for a protracted period of time, especially in a hostile and uncooperative community. The field worker should ar- range to have visits made to him by an experienced and understanding person who can, among other things, reinforce the approved image of the worker's developing self. Trips away from the field are also salutary and need to be recommended whenever circumstances permit. So important is this move to a more congenial society that the field workers of the Sum- mer Institute of Linguistics stay no longer than six months even when working in teams.