A Sociolinguistic Mismatch: Central Alaskan Yup'Iks and the Legal System

A Sociolinguistic Mismatch: Central Alaskan Yup'Iks and the Legal System

Phyllis Morrow

Morrow, Phyllis. (Summer 1993). "A Sociolinguistic Mismatch: Central Alaskan Yup'Iks and the Legal System." Alaska Justice Forum 10(2): 1, 5-8. Legal professionals working in the Central Alaskan Yup'ik (Eskimo) region of southwest Alaska commonly observe relatively high rates of confession and guilty pleas among Yup'ik clients. This article describes data from 1991 field research in Bethel, Alaska which reveal significant differences between prevailing legal and Yup'ik sociolinguistic norms that appear to contribute to this pattern, and recommends measures to alleviate problems of miscommunication which may result in unequal justice. A shorter version of this article was published as "The Language of Justice: Central Alaskan Yup'iks and the Legal System" in Criminal Justice 9(2), Summer 1994 [American Bar Association, Section of Criminal Justice].

Legal professionals working in the Central Alaskan Yup'ik region of Southwest Alaska commonly observe relatively high rates of confession and guilty pleas among Yup'ik clients. In 1991, a research team which included a trial lawyer (Galen Paine, Public Defender's Office, Sitka), a cultural anthropologist (Phyllis Morrow, University of Alaska Fairbanks), and a linguist (Betty Harnum, First Languages Commissioner of the Northwest Territories, Canada) isolated a number of cultural and linguistic factors which contribute to this pattern. This research reveals significant differences between prevailing legal and Yup'ik sociolinguistic norms, shows how miscommunication commonly builds in this setting, and suggests that when indigenous people like the Yup'iks find themselves enmeshed in the conventions of EuroAmerican legal institutions, unequal justice is likely to result.

The data for this study were gathered through observation, interviews, and linguistic analysis of courtroom discourse in Bethel, Alaska. For two months, researchers observed open court proceedings and had access to audiotapes of these proceedings. The majority of the data came from routine interactions such as plea entry or change, jury voir dire, and third party custodian assignment and from a few trials conducted during this period. While most of those undergoing proceedings were Yup'ik, as is consistent with the demography, non-Yup'iks ("Kass'aqs," as people originating in the Lower 48 and Europe are locally known) also appeared as defendants, jurors, and custodians, allowing for a comparative perspective. Interviews were conducted with both Yup'iks and Kass'aqs within the court system and related agencies. In addition, an informal sense of the concerns of inmates and their perceptions of the legal system was obtained by holding a series of workshops on legal procedures at the regional correctional center.

Analysis of these data revealed certain cross-cultural miscommunication patterns that are particularly critical in the administration of justice. The patterns centered around the process of interrogation, where differences in cultural expectations concerning questions and answers were tied to different expectations about conflict resolution. In particular, lawyers, judges, and law enforcement personnel, following the norms of legal discourse, tend to structure frameworks for questioning that cue a compliance response from many Yup'iks.

This compliance pattern was strong among Yup'iks who had little day-to-day interaction with Western bureaucratic systems, was evident in even the least inherently coercive court routines, such as the voir dire, and contrasted in significant ways with the responses of Kass'aqs and other Yup'iks who had more extensive daily involvement with EuroAmerican institutions, such as the workplace. While virtually all individuals, both Yup'iks and legal professionals, regardless of their familiarity with cross-cultural settings, employed a variety of communicative strategies in an attempt to repair the more obvious miscommunications, their strategies were often unsuccessful. This was partially because speakers differed in terms of language use and degree of understanding of the legal system, but also because Yup'ik and EuroAmerican strategies were based on fundamentally different approaches to the management of speech and interpersonal relationships. In fact, attempts on both sides to repair miscommunication often merely compounded it.

It is a clue to Yup'ik perceptions that the court is called "a place to be made to talk" (qanercetaarvik), rather than, for example, "a place where one brings problems for resolution" or "a place where justice is administered to wrongdoers." Clearly, the courts are, to some extent, intended to be intimidating: the solemnity of judicial discourse and garb and the formality of proceedings are meant to convey a sense of seriousness. Nonetheless, if a high proportion of Yup'iks feel coerced beyond this intended level, and if their response to this perceived coercion is to respond more compliantly to questioning than do other groups, then there are serious implications for justice. First, and of most obvious interest to legal professionals, such a communicative interplay can affect legal outcomes by increasing rates of confession and guilty pleas (and, by extension, rates of conviction) and by affecting rates of those excused from serving on juries. Given the present data, it seems likely that such imbalances occur. Second, and at another level, the courtroom can be seen as both a microcosm of intersocietal conflicts and a setting where such tensions are exacerbated. This is of critical importance to legal practitioners because it calls into question the basic efficacy of the legal system in bilingual/bicultural settings, which are becoming increasingly common throughout the country.

The remainder of this report will briefly summarize the data on which these conclusions are based. First, general attitudes towards speech and conflict resolution among Yup'iks will be compared with corresponding EuroAmerican attitudes as they are embodied in the legal system. It should be emphasized, at the outset, that these are broad generalizations, and that individual speech norms and attitudes are dependent on age, gender, education, cross-cultural experience, and other social factors that make interactions even more complex. Nonetheless, the following are widely influential sociolinguistic features, both reported and observed. Second, several examples will be elaborated to show how these differences influence courtroom interactions. Finally, the implications of these situations for the justice system will be discussed.

Attitudes Towards Speech and Conflict Resolution

First, basic differences exist between Yup'ik and legal attitudes towards quantity and effect of speech. These contrasts (some of which are similar to observations made by Ronald and Suzanne Scollon about Athabascan interethnic communications) are summarized below:

Comparative Speech Norms Affecting Legal Interactions


  • Speakers who elaborate their thoughts are generally admired as eloquent and precise.
  • If speech is judged to be empty, presumptuous, or negative, this reflects poorly on the speaker, but not on the subject of speech.
  • Analysis and specification of meaning are enlightening; quantification and measurement are precise indicators of meaning and valued forms of expression. (For a wider discussion of legal outcomes in relation to plaintiffs' structuring of their narratives in terms of "rules" vs. "relationships," see John Conley and William O'Barr, Rules vs. Relationships: The Ethnography of Legal Discourse. Chicago: University of Chicago Press, 1990.)
  • It is possible to offer an opinion about the motivations of others.
  • If another speaker's statements are untrue, or if one disagrees with his/her assessment of a situation, and if opposing views are not aired, then untruths may prevail.


  • Speakers who are conservative with their words are admired as thoughtful and careful.
  • Words can be self-actualizing; they influence the subject of discussion (as well as possibly reflecting poor judgment on the part of the speaker); i.e., negative discussion may make negative events happen, while positive wording can result in positive outcomes.
  • Analysis can lead to confusion; specification tends to pin down meanings too exactly and is therefore limiting and potentially incorrect; quantification may distract from substantive understanding of issues.
  • It is not prudent (or ultimately possible) to speak knowledgeably about another's motivations.
  • Even if one does not agree with others' statements or assessments, it is important to respect their views, since truth is not the province of any single viewpoint. Truth will prevail; it is best not to openly contradict others.

In one example of a mismatch between such expectations, Yup'ik prospective jurors in Bethel seemed more reluctant than EuroAmericans to respond directly to quantitatively-framed questions regarding the level of alcohol intoxication: "You think you'd be able to evaluate yourself . . . where you would have been on a scale of say one to ten, where one is stone cold sober and ten is passed out, if you were able to say when you get to a certain point, a six or a seven or something like that?" To this request for assessment, a EuroAmerican juror simply answered, "Yes," while a Native juror responded, "I wouldn't - I've never thought about it in those terms. If you could be more specific on what you want me to say, then I could tell you, but I've never thought about it."

Second, in contrast with western interactional norms, the primary flow of information in Yup'ik society is not through direct questions and answers. Requests and questions are generally avoided because they put others in a position where they are expected to comply (even if this might be difficult or impossible due to unforeseen conditions). When direct requests and questions are used, it is primarily by those in positions which are socially assigned respect - for example, from elders to youths - - and with whom compliance is the norm. Instead of asking others to tell about themselves or to do things, social relationships among Yup'iks are largely managed by anticipating each others' interests, offering indirect indications of one's needs and/or intentions, and allowing others a range of interpretations of meaning in any situation. Obviously, among Yup'iks, as in all speech communities, individuals can express subtle shades of meaning and relationship to make it more likely that others will comply with implied requests. However, the more formal or distant the relationship between speakers, the more care is taken to be indirect. Thus, the basic tendency to comply with questions/requests and offer listeners a range of interpretations is accentuated in cross-cultural and/or socially uncomfortable situations. Clearly, this can make legal interactions confusing for all parties involved.

The grammatical structure of the Yup'ik language itself makes it easy for a range of social and philosophical meanings to be implied or left open to the listener's interpretation. These features of Yup'ik also characterize the English dialects which have arisen in Yup'ik communities. Following are examples of several structures that are often problematic in legal settings:

  • A range of meaning from "allowing" to "compelling" is expressed in one form. In English, this form is expressed as "let," as in "He let me get into his car." Contextual clues (e.g., if the speaker is testifying in a sexual assault case) would point to the interpretation of force in this example.
  • A range of meaning from intentional to unintentional error is expressed in a single form. Consequently, in Yup'ik English, "joking" and "lying" may be used interchangeably.
  • A predisposition or tendency may be expressed as a desire ("wants" and "tends to" are a single form): "She wants to run away."
  • Qualifiers are used to soften assertions, even when the speaker feels certain of the information being conveyed. In Yup'ik English, this appears as frequent use of, e.g., "maybe" and "sometimes." When used along with "always" or "never," this strategy makes the statement stronger to Yup'iks, but more ambivalent-sounding to English monolinguals: "He sometimes never lets me go to Bethel" suggests that he does not allow her to go to Bethel.
  • Yup'iks commonly answer questions intended to elicit definitive opinions or assessments of others' actions with "I don't know" (expressed verbally or with a shrug), in response to the inappropriateness of and discomfiture occasioned by such questions.

The speech norms described above are closely related to norms of conflict resolution in the two cultural settings. EuroAmerican speech conventions are consistent with a class-based society and state power. The state compels individual compliance with laws and imposes sanctions for non-compliance. The state's hierarchical prerogative allows citizens to be questioned directly about their actions. The legal system defends individuals by allowing them to remain in an essentially adversarial position vis a vis interrogation; that is, they do not have to admit guilt under direct questioning - the state has the burden of proof. For the Yup'iks, social control consisted, and to a large degree continues to consist, of the actual or threatened denial of reciprocity - understood here as exchange of goods and services structured by frameworks of mutual obligation. In classless societies such as that of the Yup'iks prior to European colonization, according to Jane Fishburne Collier in Marriage and Inequality in Classless Societies (Stanford, CA: Stanford University Press, 1988, pp. 224-225), "denial of reciprocity is . . . the equivalent of state power." Formerly, ostracism and abandonment were the denials of reciprocity invoked in cases of severe transgression; symbolic demonstrations that group support was being denied were also effective in lieu of these more direct punishments.

Since mutual interdependence in the Arctic is so clearly essential to survival, these were serious consequences, and conflict was prevented from escalating to this point when possible. Prevention took two forms. First was a belief that wrongdoers would eventually get what they deserved (e.g., their denial of harmony would come back to them in the form of misfortune, such as lack of hunting success), just as proper actions brought good to self and community. Human revenge was generally unnecessary, although if retribution did not occur naturally, humans did take action. Second was the idea that positive speech - describing harmonious relations, admonishing individuals of the consequences of their behavior, confession of misdeeds, and emphasis on future good behavior rather than past misconduct - both prevented further wrongdoing and served to resolve conflicts: ". . . [T]he purpose of talking to an offender and listening to that person confess transgressions is to reintegrate both parties in the conflict into the normal functioning of the community" (A. Fienup-Riordan "Eskimo Law and Order," in Eskimo Essays, Rutgers University Press, New Jersey, 1990: p. 213). The speech conventions described earlier also facilitated smoother social relations; if one did not give offense, one invited reciprocity. Here, the importance of leaving room for individual differences, not assuming others' motivations, and not putting others into a position where they might have to refuse a request become very clear. These norms continue to shape social interaction and are reflected in the speech conventions described above.

It should be made clear that behavioral conventions are not synonymous with character: Yup'ik compliance strategies do not suggest that Yup'iks are compliant people any more than EuroAmerican strategies of direct questioning imply that EuroAmericans have an inherently aggressive character. Within each group's speech community, these are merely routine ways of interacting, understood within a familiar set of expectations about what will and will not happen, depending on how one says what one says. What is important here is that conventions affect interactions cross-culturally, because parties tend to misconstrue each other's intent.

Communications in the Court

The following examples briefly illustrate Yup'ik strategies of compliance in several types of legal encounters.

Compliance and Confession

To determine whether subjective impressions of a high confession rate were true, researcher Galen Paine conducted a preliminary analysis of 155 reportable cases in Bethel. Of these, 59 per cent of those who were questioned confessed to the crime under investigation. Paine characterized 49 per cent of those confessions as "truthful;" that is, the person volunteered or elaborated on incriminating information, according to court records. The degree to which the individual elaborated seemed independent of the degree of coercion applied by an interrogator. This sample was contrasted with 118 reportable cases from Sitka, where the suspects were primarily non-Native (a few were Tlingit). Only 51 per cent of this sample confessed to the crime at hand; more strikingly, however, only 32 per cent volunteered incriminating details.

Clearly, this is a small sample, and the results are merely suggestive of the conclusions that the confession rate in the Bethel area is comparatively high and also contains significantly more voluntary confessions that exhibit incriminating detail. More extensive studies are needed; however, these results certainly do not contradict the prevailing impressions of our respondents.

Interview respondents attributed these characteristics of Yup'ik confession to the fact that "Yup'iks are honest," and to the sense that individuals hoped confession would expedite a resumption of normal social relationships. Both inmates and attorneys also noted that repeat offenders tended to act more self-protectively, and were somewhat less likely to confess; they commented that the system ironically taught people to lie.

Compliance in Scripted Interchanges

In more routine types of questioning the researchers observed another form of compliance. Here, defendants almost always answered the court's scripted questions with predictably "correct" responses. For example, the question "Do you understand?" was answered affirmatively, even when subsequent discussions or events made it apparent that the person did not understand several of the points to which he had agreed. When defendants did give an unpredicted answer, they tended to withdraw it quickly when it became obvious that the response attracted attention. The following segment from one change of plea serves to illustrate:

Compliance in the Voir Dire

In the voir dire procedures which were observed, there were clear contrasts between question-response sequences with people whose interactive style reflected more local Yup'ik ("Y") patterns, and those who accommodated to wider English ("E") norms - generally those who had been raised more biculturally and held bureaucratic jobs in Bethel. Tapes of one voir dire sequence in the judge's chambers, for example, indicated that although each prospective juror stated the same reasons for being excused (that he or she knew some of the people involved and could not be fair), Y and E individuals were treated differently. Y jurors were questioned up to five times as long as E jurors and expressed considerably more discomfort. Research suggests that differential treatment was related to differences in jurors' communicative styles.

E jurors stated their biases directly: "I feel I already have my mind made up . . . against the defendant." Y jurors, by contrast, made their assertions indirectly. They opened with statements such as: "I don't think I could, you know, be fair." Interpreting their qualified statement as hesitancy, attorneys pressed them for more definitive statements. In response, the prospective jurors resisted reformulating their positions more directly. Instead, they addressed the issue of fairness: "I think I'd just listen to the case and try to be fair for both sides but I'd be, you know - I mean - knowing these people. . . ." Another strategy was to shift the topic to their general willingness to comply: "I mean I've been willing to help all this time in a big city," or "I'd be glad to serve any other case, but. . . ."

The Y jurors appeared to face a compliance dilemma: To be excused they had to state unequivocally that they could not be fair, yet this would seem to demonstrate a stance of noncompliance with the behavior expected of jurors (and, perhaps, those of a good person in general). Attorneys, faced with what appeared to be inconsistent or equivocal responses, tended to pursue questioning until the Y jurors reluctantly expressed their biases directly or became so uncomfortable that they ceased to speak at all. Persistent questioning is a strategy commonly used by attorneys to clarify ambivalence. In this context, however, the prospective jurors were not expressing ambivalence about their biases in the case; rather, they were expressing a cultural preference for not stating those biases in direct, unequivocal terms. Since attorneys misunderstood the sources of these ambivalent responses, their interrogative strategies were largely unproductive. Ironically, then, the people who seemed most distressed by continuous questioning were thus the ones most persistently interrogated.

Some Implications and Recommendations

When speakers operating on two differing sets of assumptions interact in the powerful settings of the law, there are a number of consequences. First, the tendency to comply with interrogation leads to ready confessions and more frequent use of the no-contest plea. If more cases were brought to trial, it seems logical that the conviction and incarceration rate, as well as the length of sentences, would decrease. While this would increase the burden on the already overworked court system, it would help to equalize treatment under the law, especially in cases where elaborated confession is more the result of compliance than of guilt. Attorneys reported many such situations, in which the defendant did not remember committing a crime but confessed when told that another person said he had committed it.

Second, the legal system unwittingly undercuts basic Yup'ik cultural assumptions about human relations to the degree that it rewards "lying" and direct, unqualified statements about the motivations of others.

The following recommendations might help to alleviate these problems:

1) Both legal professionals and the general public can be generally educated about cultural differences and their potentially profound effects. For the Yup'ik public, this might lead to discussions about the different consequences of confession within the Native community versus within the legal system; within the legal profession, alternative strategies for eliciting and communicating information might be developed.

2) Specific legal education for the public would improve knowledge about the legal system. Mock trials with discussion of legal processes have proven effective in both high schools (notably at Mt. Edgecumbe) and correctional centers. Cultural issues can also be addressed in this context.

3) Training and regularly using interpreters would facilitate understanding in the bilingual courtroom, even where no participants are monolingual in the Native language. Although court personnel indicated that only 10 per cent of those undergoing legal procedures needed interpreters, virtually all Yup'iks interviewed agreed that they would understand proceedings better if they were conducted bilingually. Clearly, encouragement for bilingual-bicultural individuals to enter the legal profession in all capacities would also be desirable. At present, understanding is often a lower priority than the establishment of an unambiguous court record and efficient management of cases. These measures could help to shift this priority.

While the conflicts discussed here are specific to the interaction of Yup'iks and the law, the general cross-cultural situation is similar in other parts of the state, and the above recommendations are largely generalizable.

Phyllis Morrow is associate professor of anthropology at the University of Alaska Fairbanks. This article reflects research conducted under a grant from the National Science Foundation. For further information on the project the author may be contacted.