Biographical Memoirs: Volume 71 (1997)

January 26, 1884-February 4, 1939

BY REGNA DARNELL AND JUDITH T. IRVINE

AMONG THE ANTHROPOLOGISTS trained by Franz Boas in the early decades of the twentieth century Edward Sapir alone was regularly acknowledged by his peers as a genius. The only professionally trained linguist among Boas's students, and gifted with intuitive insight into grammatical patterning and historical relationships of linguistic families, Sapir contributed seminally to general linguistic theory, Amerindian linguistics, and Indo-European linguistics. He also made important anthropological contributions in ethnology, culture theory, and cultural psychology. A prolific fieldworker as well as theorist, Sapir recorded for posterity thirty-nine different Amerindian languages, often working with the last living speaker. Alongside his linguistic investigations he gathered ethnographic information and transcribed indigenous-language folklore texts. He was a humanist as well as linguist and anthropologist, composing music and publishing poetry and literary criticism. For his successors in a range of disciplines he continues to exemplify the study of meaning and expressive form across linguistic and cultural boundaries.

Although Sapir was born in Lauenberg, Pomerania

(Prussia), in what is now Lebork, Poland, his parents, Jacob David and Eva Seagal Sapir, were Lithuanian Jews. Sapir undoubtedly learned German as a child, but the language of his home was Yiddish; he read Hebrew with his father, a cantor, beginning when he was seven or eight. Jacob Sapir preferred music to theology, however, and the family's daily life was not intensely orthodox in religious observance.

The family moved several times during Sapir's early childhood. He began kindergarten in Liverpool, England, while Jacob preceded his wife and children to America, obtaining a position in Richmond, Virginia, in 1890. Shortly after the move to the United States Sapir's younger brother Max died of typhoid, and Jacob's career declined through a series of short-lived appointments. The family took root on the Lower East Side of New York City when Edward was ten. Eva Sapir ran a small notions shop to support herself and her remaining son; she and Jacob divorced sometime after 1910.

When Sapir was fourteen he won a Pulitzer scholarship for four years at the prestigious Horace Mann High School. He declined it in favor of a local high school and used the scholarship for his undergraduate education at Columbia University. He was one of the bright stars among the immigrant children of the city, and higher education was his prize.

Entering Columbia in 1901, Sapir concentrated on Germanic philology while gaining formal training in Indo-European linguistics. He received his B.A. in German in 1904, having taken only three years to complete the four-year program. In 1905 he received his M.A., also in German. He took two more years of courses in anthropology and German, receiving his Ph.D. in anthropology in 1909 with a dissertation on the Takelma language of southwestern Oregon.

Languages were Sapir's forte from the beginning. Since

Columbia had no department of linguistics as such, Germanics was the field of choice for a student interested in linguistic science. After Sapir met Franz Boas, however, he was inspired by the urgency of the need to record endangered Amerindian languages before they were lost forever. To apply the methods of comparative Indo-European to unwritten aboriginal languages was, for him, an obvious step. His interest in linguistic theory went far beyond that of Boas, a self-taught linguist who acknowledged his pupil's intellectual leadership in linguistics while Sapir was still a graduate student.

The transition from Germanics to anthropology was a smooth one. Sapir's M.A. thesis on Herder's theory of the origin of language, by including Eskimo examples, already reflected the influence of Boas. At this time, as in later years, Sapir defended the functional equivalence of all human languages, explicitly including those of "primitive" peoples. But his real apprenticeship as a field linguist, in the anthropological tradition, began in 1905 when Boas sent him to the Yakima Reservation in Washington to work on Wasco and Wishram Chinook. There were many languages begging for description. In 1906 Sapir returned to the field for his dissertation research, working on Takelma and Chasta Costa at Siletz Reservation in Oregon.

Sapir's first professional appointment, in 1907, was as a research assistant at the University of California, Berkeley, where fellow Boas student Alfred Kroeber had a mandate to map the enormous cultural and linguistic diversity of the state. In a single year Sapir studied three dialects of Yana and worked briefly on Kato. But Kroeber was more interested in surface description that would classify related languages than in the careful grammatical analysis Sapir thought

should produce a dictionary, grammar, and texts for each language studied. California did not continue the appointment after its first year.

Sapir moved to the University of Pennsylvania in 1908 to take up a Harrison fellowship, which involved teaching as well as research through the University Museum. With his ethnologist colleague Frank Speck, another former Boas student, Sapir worked on Catawba. In the 1909 field season Sapir and his student John Alden Mason began fieldwork with Uintah Ute in Utah. They planned a long-term study of Ute language and culture, but their project was not funded by the museum.

Remaining in Philadelphia in 1910, Sapir began studying Southern Paiute, a language closely related to Ute, with Tony Tillohash, a student from the nearby Carlisle Indian School. It was a fortunate collaboration: Tillohash's ability to analyze his native language meshed with Sapir's intuitions to produce what has sometimes been called the most beautiful grammatical description ever written of an Amerindian language. Sapir worked briefly on Hopi with another Carlisle student but abandoned it in favor of his work with Tillohash, choosing the ideal linguistic informant over the language as such.

At the age of twenty-six Sapir obtained a plum position in the expanding Boasian network of professional anthropology in North America. He served from 1910 to 1925 as the first chief ethnologist of the Division of Anthropology in the Geological Survey of Canada, Department of Mines. As Canada's paramount anthropologist he quickly developed a research and publication program and a national museum focusing on the aboriginal peoples of the dominion.

With a wide research field to cover Sapir hired several Boas-trained researchers and alternated his own research

program between intensive work with Nootka on Vancouver Island and survey fieldwork among a variety of northeastern languages spoken within easy range of Ottawa. Although he was able to make only two field trips to the Nootka area before the First World War dried up research funds and administrative responsibility made summer fieldwork more difficult, Sapir worked with speakers of various northwest coast languages when their speakers visited Ottawa on tribal business. In 1922, a few years after the end of the war, he was able to return to a brief stint of in situ fieldwork for a study of Sarcee, an Athabaskan language, in Alberta. The following year he pursued the Athabaskan research with Kutchin and Ingalik, Athabaskan languages of northern Canada, since some speakers of these languages happened to be living not far away at Camp Red Cloud, Pennsylvania.

The Canadian work was interrupted only once, when Kroeber invited Sapir back to California to work with a “wild" Indian, the last speaker of Yahi, a Yana language. Using his knowledge of other Yana varieties studied years before, Sapir spent the summer of 1915 recording Ishi's unique knowledge of his language and culture.

The later Ottawa years were depressing ones, on personal as well as professional grounds. Sapir was a pacifist during the First World War and keenly felt his position as an immigrant to North America. Florence Delson Sapir, whom he had married in 1910, suffered from a series of mental and physical ailments until her death of a lung abscess in 1924. Sapir's mother came to help with the three children. As for his research activities, even after the end of the war the research portion of the division's work did not recover enough funding to restore its original grandeur.

These were years of intense introspection for Sapir. He wrote poetry and literary criticism, dabbled in psychology, and composed music. Largely prevented from carrying out

new fieldwork and increasingly frustrated by his inability to write up his accumulated materials on the myriad languages he already had studied in the field, he turned to more general linguistic questions and to the theory of culture, society, and the individual. Although his own intellectual activities had lost none of their vigor, he felt isolated in Ottawa and lamented the absence of a university affiliation with the chance to train his own students.

In 1925 Sapir was called to the University of Chicago, which had already assembled a stellar faculty. His appointment was to a joint Department of Sociology and Anthropology, which split in 1929. Since the "Chicago School" of sociology was the most prestigious and professional variety of social science in North America at the time, the new position placed Sapir at the center of a network of interdisciplinary scholarship, much of it sponsored by the Rockefeller Foundation. After the years of perceived isolation in Ottawa, Sapir thrived on the intellectual excitement of Chicago in the late 1920s. He eagerly joined the interdisciplinary conference circuit, becoming the man of words who enabled colleagues from sociology and psychology/psychiatry to understand the common links of their work. His collaboration with interactional psychiatrist Harry Stack Sullivan and political scientist Harold D. Lasswell is particularly notable. Because he was teaching in the social sciences Sapir found himself thinking a great deal about culture, psychology, and social science methodology. Still, in this period he did not abandon his linguistic work, even managing to make field trips to study Navajo and Hupa.

It was shortly after his arrival in Chicago that Sapir renewed an acquaintance with Jean McClenaghan, then a social work student on a practicum at the Chicago Institute for Juvenile Research. The couple were married in 1927 and had two children.

In 1931 Sapir followed Rockefeller funding to Yale University. As Sterling professor of anthropology and linguistics he was expected to bring interdisciplinary research to the Graduate Division of the university, heading a new department of anthropology and drawing social science research together into a single coherent research program. With colleagues at the Institute of Human Relations he was to offer a seminar on "the impact of culture on personality," supported by the Rockefeller Foundation. He was also to serve in a newly independent graduate department of linguistics. For the first time he found intellectually congenial colleagues in linguistic theory and Indo-European studies. A cadre of his Chicago graduate students in linguistics moved to Yale with him, constituting the first Yale school of linguistics (the second one coalesced around Leonard Bloomfield in the 1940s).

These utopian plans were undermined by local academic politics, especially by vested interests in sociology, by the economic effects of the Depression, and by currents of anti-Semitism at Yale. Sapir was overextended and unhappy. Outside Yale he continued with his interdisciplinary activities; within it he focused on his own teaching in anthropology and linguistics.

In 1937, while teaching at the Linguistic Society of America Summer Institute at Ann Arbor, Michigan, Sapir suffered his first heart attack. A sabbatical trip to China in 1937-38 had to be cancelled because of his health. Although he returned to teaching in the fall of 1938, he had not recovered his strength. He died early in 1939 at the age of fifty-five.

Sapir's first synthetic works were part of the formalization of the Boasian paradigm. In 1916 his Time Perspective in

Aboriginal American Culture: A Study in Method laid out the method of historical inference implicit in the Boasian reconstruction of the history of cultures and languages. (At the time, direct archaeological evidence of American prehistory was scanty, and there were no consistent standards for its interpretation until the Pecos Conference a full decade later; indirect evidence, such as might be provided by linguistics and ethnology, was therefore crucial.) Drawing on linguistic examples from a remarkable range of cases, Sapir in Time Perspective distinguished methodologically between the properties of language and culture for historical reconstruction. Sound change in language, unlike the other parts of culture, he argued, retained traces of the past historical relationships of languages. In consequence, genetic relationships could be discerned and distinguished from other kinds of relationships by the application of methods used in Indo-European historical linguistics, even in the absence of written records. Sapir's treatise remained the ethnologist's guide to historical method for a generation and still repays careful attention to the forms of his logic.

In 1921 Sapir published Language: An Introduction to the Study of Speech, the only book he completed during his lifetime. He included written and unwritten languages on an equal footing, marvelling at the precision and beauty of grammatical forms and structural typologies. This was Sapir the linguist writing at his most lyrical and persuasive. The book was directed at an educated general audience, but its broad canvas and penetrating vision of linguistic form, as well as its treatment of specific topics, have greatly influenced professional linguists ever since. The discussion of "drift," for example, remains fundamental to linguistic theory about processes of language change.

Also in 1921 Sapir published a one-page summary of his six-unit classification of American Indian languages, based

on a paper read to the American Association for the Advancement of Science. Although the 1929 version of this classification is better known and is accompanied by considerable justification, including a medial classification of twenty-three units acceptable even to conservatives among Amerindian linguists, the 1921 version was essentially complete. It was based on the comparative work Sapir and his colleagues had done over the past two decades. Although Sapir himself saw the classification as a series of working hypotheses, many anthropologists promptly reified its categories, latching onto the six-unit classification as an easy guide to tribal relationships.

The most daring of the proposals made by Sapir in this period involved linking Athabaskan to Haida and Tlingit to form Na-dene and then linking Na-dene, largely on the basis of its tonal structure, to Sino-Tibetan. By the 1930's, however, when Sapir moved to Yale, his colleagues in linguistics were skeptical of such speculative large-scale genetic hypotheses, and the anthropologists were no longer in dire need of historical models from linguistics (if only because of the emergence of reliable dating methods in prehistoric archaeology). During the Yale years Sapir paid less attention to the six-unit classification, returning instead to linguistic theory and to specific linguistic problems both within and beyond the Americanist field, including studies in African, Semitic, and Indo-European linguistics.

Some of Sapir's most famous contributions to linguistic theory lie in phonology, the study of sound systems. In 1925 the inaugural issue of Language-the journal of the Linguistic Society of America, of which Sapir was a crucial founder-carried his paper, "Sound Patterns in Language," which defined the concept of the phoneme in terms of significant relationships among sounds, rather than their objective qualities. In 1933 he followed up this pattern-ori-

ented argument in discussing the phoneme's "psychological reality," that is, the intuitions of Amerindian language speakers for their native language's phonological system. The level of generalization implicit in Sapir's distinction between phonetics and phonology in these papers, which revolutionized American linguistics, was derived from fieldwork with aboriginal languages independently of parallel work on phonemic models by the Prague School of linguists in Europe. A late (1938) paper of Sapir's on glottalized continuants pursued these phonological themes and is significant for its use of evidence from Amerindian languages alongside Indo-European data.

Sapir is also especially noted for his dynamic conception of grammar. His analysis of the grammar of Southern Paiute, together with his student Stanley Newman's grammar of Yokuts, stand as exemplars of the "process grammar," an important though discontinuous precursor of contemporary generative theories. What intervened was the school of linguistics associated with Leonard Bloomfield, Sapir's younger colleague at Yale. Sapir's conception of grammatical process and his interest in the study of meaning as integral to the theory of grammar contrast sharply with the work of the Bloomfieldians.

Sapir's discussions of the role of meaning in grammatical form and the relationships of these to the use of language in formulating and conveying ideas have been taken as his contribution to what is often called the Sapir-Whorf hypothesis. In fact the hypothesis was developed largely by his student Benjamin Lee Whorf after his mentor's death. But there are certainly intimations in Sapir's own writing of the way in which habitual thought might be influenced, if not determined, by linguistic structures.

There is almost no important topic in linguistics or its allied disciplines to which Sapir did not contribute. Some

of his interests, it is true, no longer command widespread scholarly attention, such as the construction of an international language. Others, such as his work on sociolinguistic variation in Yana, have been rediscovered by modern scholars who emphasize these topics more than Sapir himself did. Taken as a whole, however, the range of Sapir's concerns significantly shaped the outlines of American linguistics for later generations.

Although Sapir's reputation in the decades following his death has rested more upon his contributions to linguistics than upon his role in cultural anthropology, during his lifetime he was known as an important ethnologist and cultural theorist as well. In 1916 after the publication of Time Perspective—an essay that includes explorations of the diachronic implications of ethnological phenomena, on analogy with language—he embarked on a consideration of theoretical problems in the concept of culture. These interests were to occupy him increasingly during the rest of his professional life. His 1917 debate with Kroeber on the "superorganic," a debate in which Sapir challenged Kroeber's assumptions about anthropological epistemology and the role of individual achievement and experience in cultural systems, was only the first of many discussions of these themes.

Sapir's conception of culture and anthropological method was always influenced by his work in linguistics. Language was, for him, the cultural phenomenon par excellence. It offered the prime example of cultural difference and cultural systematicity; it provided the ethnographer with the terminological key to native concepts; and it suggested to its speakers the configurations of readily expressible ideas. But Sapir's thinking about culture drew significantly, as well, on his interests in psychology and psychiatry, especially Jung's

writings on personality, Koffka's Gestalt psychology, and Sullivan's interactional psychotherapy.

One of the problems that most interested Sapir was the tension between the anthropologist's concern with abstracting cultural patterns from observable behavior and the individual participant's personal biography and subjective experience. In contrast to many other anthropologists of the time Sapir emphasized intracultural variability, disagreement, and individual agency. He distinguished carefully between, on the one hand, subjective meanings and experience, and, on the other, the public symbols and social conventions prescribing the forms a person's behavior takes. Although much interested in the relationships between culture and personality, Sapir criticized approaches which, in his view, failed to distinguish collective and individual levels of analysis, confusing conventional patterns of behavior with the personality patterns of actual individuals. Late in his life, influenced by his collaboration with Harry Stack Sullivan, Sapir began to look to the analysis of social interaction as the locus of cultural dynamics.

Sapir's writings on culture have sometimes been seen as falling into an extreme methodological individualism, but this view distorts his position. He was equally interested in cultural configurations and in the ways an individual's experience is dependent on social setting. The problem was how a theory of culture could accommodate both its individual and its social sides. Since his peers, Kroeber especially, seemed to give priority to the social, Sapir's writings often emphasized the individual.

During his lifetime his contribution to cultural theory took the form of a series of essays. Although he planned to write a book on "the psychology of culture," based on his graduate lecture course of that title and the Rockefeller seminar at Yale, he did not live to complete it. A manu-

script for the book was finally reconstructed posthumously from students' lecture notes and published in 1993.

Sapir's scholarly reputation is easily documented in the official honors accorded him. His positions at Chicago and Yale, an honorary degree from Columbia, elections to the presidencies of the American Anthropological Association and the Linguistic Society of America, his membership in the National Academy of Sciences, the memorial volume (originally planned as a festschrift) published shortly after his death, and many other honors are evidence of the scholarly esteem in which his colleagues held him. Their respect was also personal. As his student David Mandelbaum wrote in an obituary of Sapir published in 1941, "He was more than an inspired scholar, he was an inspiring person. Listening to him was a lucid adventure in the field of ideas; one came forth exhilarated, more than oneself. . . . An eminent psychiatrist recently remarked that Sapir was an intoxicating man. That he was."

Yet despite the force of his personality and the importance of his contributions, there is no "Sapir school" in either of the major disciplines to which his work was foundational. No single one of Sapir's students pursued all of the disciplines or topics that consumed his interests over the course of his career. His untimely death in 1939 left them without a mentor at a time when a world depression and then a world war took priority over scholarly concerns. After the war there were many changes in the academic scene. As linguistics became an autonomous discipline its ties to anthropology weakened in a number of ways. Not all anthropologists were expected to be linguists some of the time, and linguists were doing specialized work beyond the capabilities of scholars lacking very intensive training. An-

thropology, meanwhile, expanded both in geographical area and in size of the profession. Amerindian studies could no longer be seen as the core of anthropology as they had been for the first half of the century. In the late 1940s and 1950s the "culture and personality" school associated with the work of Margaret Mead and Ruth Benedict in anthropology and the structuralist school of Leonard Bloomfield and his students in linguistics took positions opposed to Sapir's and temporarily dominated the fields in which he had principally worked.

Nonetheless, the continuities were there, and they have emerged in the responsiveness to Sapir common among students of the students of his students. In recent years scholars in both linguistics and anthropology have rediscovered the continuing relevance of his work. The centenary of Sapir's birth in 1984 produced a spate of Sapir scholarship, including several conferences and collections of papers, a biography by Regna Darnell, a reprinting of David Mandelbaum's (1949) Selected Writings of Edward Sapir in paperback, a reconstruction of The Psychology of Culture by Judith T. Irvine, and a plan by Mouton de Gruyter to publish a definitive collected works in sixteen volumes (six of which have now appeared) under the general editorship of Sapir's third son Philip.

There is probably no North American linguist or anthropologist today who does not respect, even revere, the name of Edward Sapir. He set a standard for the integration of disciplines—linguistics, anthropology, psychology, and the humanities. He wrote grammars of process rather than static formalism. He treasured the study of meaning and the myriad forms in which it could be expressed. His concept of human nature and communication, which included primary research in Amerindian, African, Indo-European, and Semitic languages, was sufficiently broad to encompass any and all

human languages. These are ideas and approaches which have come full circle in the half century since Sapir's death.

1907

Herder's Ursprung der Sprache. Mod. Philol. 5:109-42.

1909

Ed. Wishram texts, together with Wasco tales and myths. Collected by Jeremiah Curtin. American Ethnological Society Publications II.

Takelma texts. Univ. Pa. Anthropol. Publ. 2(1):1-263.

1910

Yana texts. Univ. Calif. Publ. Am. Archaeol. Ethnol. 9:1-235.

1911

The problem of noun incorporation in American languages. Am. Anthropol. 13:250-82.

1913

Southern Paiute and Nahuatl, a study in Uto-Aztekan. Part I. J. Soc. Am.-ist 10:379-425.

Wiyot and Yurok, Algonkin languages of California. Am. Anthropol. 15:617-46.

1914

Notes on Chasta Costa phonology and morphology. Univ. Pa. Anthropol. Publ. 2(2):271-340.

1915

Southern Paiute and Nahuatl, a study in Uto-Aztekan. Part II. Am. Anthropol. 17:98-120, 306-28 and J. Soc. Am.-ist 11:443-88.

1916

Time Perspective in Aboriginal American Culture: A Study in Method. Canada Department of Mines, Geological Survey, Memoir 90. Anthropological Series, No. 13.

1917

The position of Yana in the Hokan stock. Univ. Calif: Publ. Am. Archaeol. Ethnol. 13:1-34.

Do we need a superorganic? Am. Anthropol. 19:441-47.

1921

Language: An Introduction to the Study of Speech. New York: Harcourt Brace.

A bird's eye view of American languages north of Mexico. Science 54:408.

1922

The Takelma language of southwestern Oregon. In Handbook of American Indian Languages, part II, pp. 1-296. Bureau of American Ethnology, Bulletin 40.

1924

Culture, genuine and spurious. Am. J. Soc. 29:401-29.

1925

Sound patterns in language. Language 1:37-51.

1927

Anthropology and sociology. In The Social Sciences and Their Interrelations, ed. W. F. Ogburn and A. Goldenweiser, pp. 97-113. Boston: Houghton Mifflin.

Speech as a personality trait. Am. J. Soc. 32:892-905.

1928

The unconscious patterning of behavior in society. In The Unconscious: A Symposium, ed. E. S. Dummer, pp. 114-42. New York: Knopf.

1929

Central and North American languages. Encycl. Br. 5:138-41.

The status of linguistics as a science. Language 5:207-14.

1930

With L. Spier. Wishram ethnography. Univ. Wash. Publ. Anthropol. 3:151-300.

Totality. Language Monograph No. 6. Linguistic Society of America.

The Southern Paiute language: Southern Paiute, a Shoshonean language; Texts of the Kaibab Paiutes and Uintah Utes; Southern Paiute dictionary. Proc. Am. Acad. Arts Sci. 65(1):1-296; (2):297-536; (3):537-730.

1931

The concept of phonetic law as tested in primitive languages by Leonard Bloomfield. In Methods in Social Science: A Case Book, ed. S. A. Rice, pp. 297-306. Chicago: University of Chicago Press.

1932

With M. Swadesh. In The Expression of the Ending-Point Relation in English, French, and German, ed. A. V. Morris. Language Monograph No. 10. Linguistic Society of America.

Cultural anthropology and psychiatry. J. Abnorm. Soc. Psychol. 27:229-42.

1933

La realite psychologique des phonemes. J. Psychol. Norm. Pathol. 30:247-65.

1934

The emergence of the concept of personality in a study of cultures. J. Soc. Psychol. 5:408-15.

1936

Internal linguistic evidence suggestive of the northern origin of the Navaho. Am. Anthropol. 38:224-35.

1937

The contribution of psychiatry to an understanding of behavior in society. Am. J. Soc. 42:862-70.

1938

Glottalized continuants in Navaho, Nootka, and Kwakiutl (with a note on Indo-European). Language 14:248-74.

Why cultural anthropology needs the psychiatrist. Psychiatry 1:7-12.

1939

With M. Swadesh. Nootka texts: Tales and Ethnological Narratives with Grammatical Notes and Lexical Materials. William Dwight Whitney Linguistic Series. Philadelphia: Linguistic Society of America.

1942

Navajo Texts, With Supplementary Texts by Harry Hoijer, ed. H. Hoijer. Philadelphia: Linguistic Society of America.

1943

With L. Spier. Notes on the culture of the Yana. Univ. Calif. Publ. Anthropol. Rec. 3:239-98.

1947

The relation of American Indian linguistics to general linguistics. Southwest. J. Anthropol. 1-14.

1949

Selected Writings of Edward Sapir in Language, Culture and Personality, ed. D. G. Mandelbaum. Berkeley: University of California Press.

1993

The Psychology of Culture: A Course of Lectures. Reconstructed and edited by J. T. Irvine. Berlin: Mouton de Gruyter.