SIL International Home

Kenneth L. Pike (1912–2000)

Talk, Thought, and Thing
The Emic Road Toward Conscious Knowledge

by Kenneth L. Pike

Previous | Next | Contents

Chapter 2

The person, as observer, is tied emically to things (and concepts) via differences, sameness, and appropriateness.

Conviction 2.1. For the native observer within a particular culture, emic units are considered to differ from one another in relation to everyday behavior.

Analogy: A boy is not a girl. A mountain is not a valley. Yesterday, viewed from now, is not today. Youth is not age. Emic units contrast. The contrast between the two vowels in bit and bet, in English, makes them emically different (called phonemically different when sounds of language are under discussion).

Source: I coined the term EMIC (in Pike 1954) from the linguistic term PHONEMIC, to apply to contrastive items of nonphonological material rather than just to phonological data. It is easy for us to differentiate an airplane from a wheelbarrow. But it is not easy for some of us to differentiate a copperhead snake from a rattler, even though it may be simple for people who are used to avoiding the both of them. They need to be seen as emically different, for survival, in some contexts.

Further Implications: Several years ago when I was lecturing to the philosophy club at the University of Texas at Arlington, a phenomenologist, Professor Lenore Langsdorf, said to me, "You sound like Immanuel Kant." I was surprised. I had not thought of that—but on checking it I found that my linking of observer to thing has some overlap with Kant in thought. Kant says (1966:70), "We can attain to a knowledge of appearances only, never to that of the things in themselves." I would agree that there is no way to completely eliminate our own observer impact on our own relation to the universe. Our categorization of elements of our universe allows us to have a partial understanding of the universe. As we categorize it, we turn it into things as if they are isolable or recurrent, even though they can never occur outside of some kind of physical or mental context. Units resulting from such categorization I call EMIC UNITS.[4] Years ago (1957), a philosopher in Spain, Ortega y Gasset, adopted Kant's viewpoint, and developed it in detail.[5]

Conviction 2.2. Any theory which we can live by must be able to grant that insiders to a system (such as the native speakers of a language) may call or treat several items or concepts as the same even though outside analysts may notice differences between them. This is EMIC SAMENESS in the face of ETIC DIVERGENCE. In some instances the insider responds so strongly to the sameness that it may take special training for him or her to recognize differences. In others, the differences may be obvious but ignored, as irrelevant to the communicative purpose of the speaker.[6]

Source: It was with astonishment, years ago, that I learned that the two p-sounds of the word paper are different in my own dialect of English—the first is aspirated (pronounced with a tiny puff of breath following it), but the second is unaspirated. The sameness was for me emic (phonemic); I had not observed the etic difference until I had studied some phonetics. On the other hand, with the concept 'house', I had no difficulty in calling a big house and a little house by the same term. The sensitivity to the etic variability or spread of applicability in the second case was much broader than in the first. The variability within sameness may be easier for the insider to recognize in relation to nonphonological items and behavior than it is within phonology.[7]

Further Implications: For reading on the importance of the distinction between EMIC and ETIC from contrastive viewpoints (with synchronic focus versus diachronic focus), see Headland, Pike, and Harris 1990.

Conviction 2.3. Emic difference is easier to detect when two somewhat similar items occur appropriately in the same position in a typical sequence of behavior, but where the two items are perceived by the insider as implying significantly different meanings or behavioral relevance. Emic contrasts—emic differences—occur in appropriate places in sequential behavior or in a systemic arrangement.

Emic units relevant to behavior occur in contrastive positions in that behavior. The sounds p and t contrast (phonemically, and hence emically) at the beginning of the words pie and tie, but the two ps of paper, although different (phonetically, and hence etically), occur in different positions in the word and are sufficiently alike to be ignored in relation to semantic relevance. On the other hand, if one says the same word pie a thousand times, a machine can register slight differences in each of the pronunciations. These differences are etic variants which do not carry semantic implications. Occasionally further problems arise, as when a person can have two emically different pronunciations of the same word. His colleagues know that they are different, but they can know that difference because the sounds are contrastive in other circumstances. He may say "I am going," rapidly, as "I'm going."[8]

Analogy: A pancake is not a hot dog; they contrast emically in our culture. But a pancake may be relatively thick or relatively thin and still be a pancake. In my youth, a pancake would have been more appropriate at breakfast, whereas a hot dog would have been more appropriate for lunch or supper.

Conviction 2.4. Both the physical form of an item or action and its social meaning or impact comprise parts of an emic unit. The observer, therefore, is involved in the contrastive categorization of the environment, even in those instances in which there is physical isomorphism between two emically contrastive units.

Analogy: A person may kill someone by hitting him with an automobile. A second person may do likewise. But if the first person did it unintentionally, that social component helps categorize the incident as an ACCIDENT; but if it was done intentionally by the second person, that would be categorized as MURDER. Courts of law, therefore, in this kind of situation, try to investigate not only the physical facts, but also the mental involvement of the actors and the observer interpretation or evaluation of intents, lying behind the observed actions. All of these are a part of the emic categorization.

Further Implications: For the necessity of both form and meaning, discussed in relation to tagmemic principles, see Pike 1982:109–17.

Conviction 2.5. I reject, therefore, as not being an interpretation of life which I can in fact live by, an analysis which treats human behavior exclusively in relation to mental components of an individual. I also reject, for my purposes, any analysis which treats all of behavior as exclusively physical. Similarly, I must reject an analysis of behavior which separates mental and physical so sharply that they fail to take account of their observable overlap when the observer acts to categorize his environment emically.

Analogy: In (1a), I draw two circles, completely separated from each other, and label the first circle 'ideas' and the second one 'physical action'. This implies two kinds of activity, but leaves them totally separate. Then, in (1b), a single circle represents only ideas, with no physical action believed in. (1c) does the opposite, with only physical action, without ideas. All of these I reject. In figure (1d), however, the partial separation plus partial overlap suggests the kind of view I hold, where self and body are in part overlapping and in part distinct.

Source: The observer names material and, in thus categorizing his environment, in part gets to know his environment; and he or she in part creates for his or her own mental and social activity the emic categorization of that environment. This viewpoint seems to me to leave better room for the SELF (or soul) to exist along with body, not as merely a part of it (see chapter 4).

Further Implications: I had supposed (without pondering deeply upon the problem) that when I wrote and published my book on phonemics (1947a) that I was doing it from a mechanistic perspective. I was startled, therefore, to receive a letter from Czechoslovakia saying that they were delighted to see my book, after having seen mechanistic material from the Yale school. I asked myself how they could have known I was not philosophically a mechanist even though I had thought that I was writing as one. I decided, later, that it must have been because I insisted so strongly, without discussing the underlying philosophical reasons for it, that the native speaker had reactions of understanding or of identity or of difference which were relevant to the methodology of analysis. See, for example, Pike 1947a:64 (and cf. 160), with 'native reaction' as a source of my premises: We want the student "to arrive at an analysis which parallels the vague or explicit observable reactions of speakers to their own sounds"; and "the observation that speakers of English have difficulty in learning to distinguish between the two [p] sounds in 'paper' leads to the conclusion that in some way the sounds are 'the same' for them."


[4] And note that such categorization may be called 'naming' as in Convictions 1:4–5.

[5] See his revision (Ortega y Gasset 1985:232). It is the thing within our lives, not the thing in itself, which he considered important. Note, also, related items on pp. 44f, 163, 173, 196f, 201f, 230.

[6] Emically, I would say, one can step into Heraclitus' river twice, but etically it will be a different river each time because, he says (Frost 1962:8), "You could not step into the same river twice, for other and yet other waters are ever flowing on"—with different molecules each time.

[7] See Conviction 4:5 below, for a discussion of WAVE.

[8] Methodological difficulties arise in applying these criteria, but they are not appropriate to the discussion here. For my earlier discussion of this methodology, see Pike 1947a.

Previous | Next | Contents