addition, even though both main possibilities are included in the description-in-progress, and even though both main possibilities are in frequent use by native speakers, one understanding and use occurs in 10–15% of cases, and one understanding and use occurs in 85–90% of cases. Curricula necessarily oversimplify material to minimize possible confusion for students (minimizing confusion is a prescriptive ideal; presenting material in its full complexity has not, to my knowledge, ever been tried, particularly at elementary levels of instruction). That kind of minimization would likely privilege the majority version of the language over minority versions of the language, as has already happened in the orthography. In the case of Hobongan, such privilege might not appear to be a huge problem, but in other languages, such as Daqan and English, in which dialect variations are major parts of life, such privilege can disadvantage thousands or millions of people, as well as provide access to the advantages of the majority language and culture. The prescriptivism that can be developed from descriptions has real consequences for individuals and minorities-within-minorities.
On the other hand, as an anonymous reviewer has pointed out, noting variations in languages can be a way to generate interest among students of language and linguistics and could therefore be used to enhance the education of students and the preservation of variants. However, even if variants are presented, they tend to be presented in contrast to a standard. This pattern might not need to be the case, but as a student and professor of linguistics at various times, I have never encountered a presentation in which variants were truly treated as options among equivalents, and my own efforts in that direction have been inadequate. Often, a description provides the standard by which variants are defined as such. There is a standard, and there are variants, and descriptions often contribute to making that distinction. For example, if I had been working primarily with one of the outlying groups of Hobongan speakers, rather than with one of the larger groups in one of the larger villages, I might have privileged, intentionally or not, a set of variations that are used by a small minority of people. A description based on that privilege could pragmatically imply that the variant described is the standard. If the description were used as the basis for curricula, the minority variant would become the standard, and a majority of students would be educated to conform to a minority variant. It is not necessarily ethically preferable to educate minority students to conform to a majority variant, or vice versa. A language description that includes some variants, noting who uses the variants and when and how, would pragmatically privilege some variations over others, even if the information in the description were purely descriptive, because the social distinctions, and the sociolinguistic markers that indicate those distinctions, would emerge from the description.
The form of a language chosen and used for education, then, becomes the form of a language preferred when language is used for writing. As a reviewer pointed out, in language descriptions, privileging some aspects of language such as syntax, over others such as analyses of various common genres, might suggest to native speakers that the privileged aspects of the language are in fact more important than the others. This privileging approaches a prescriptive-style focus on form over larger-level structures in the language which contrasts with the descriptive importance of larger-than-sentence units of language. The Hobongan have already lost several genres of chants and incantations in converting to Christianity. Their language considered as langue is important enough to warrant documentation and translation efforts, but their uses of language might not be important enough to them to maintain. Nevertheless, the Hobongan might begin to record their remaining oral histories, stories and songs and to create new histories, stories and songs.
Furthermore, traditional materials might exist in parts of the community whose language variety has not been recorded or given priority in description, but when the materials are recorded in written form, that written form would probably be what is standardized for education, which is the context in which people acquire writing. The standard form will miss some of the information that might have been available if each variant were given equivalent importance. A language description can have prescriptive consequences for any given language well into the future.
2.4 For ease of governance and commerce
Language descriptions are also used by agents of governments to determine whether people should receive legal or institutional protections for their language and, if so, what those protections should be. Language documentation is a crucial component of gaining minority rights in Indonesia because people are assumed to be Indonesian and speak BI until the people can document that they are who they say they are. Ironically, minorities must accommodate majority language and ways of doing business in order to demonstrate that they are not the majority. As with using language descriptions for developing educational materials, one variety of a language is typically privileged over other varieties, in part because the complexities of language in actual use make governance of actual people extraordinarily difficult. A description can be used, in prescriptively best-case scenarios, to protect minorities and individuals, but it can also be used to impose the prescriptive ideals of the majority-within-the-minority on all members of a community, because descriptions necessarily privilege one form of a language over other forms. The finite and descriptive necessities of language documentation can therefore justify or become government policies, again making for real consequences for real people.
3 Summary and Conclusion
How is a linguist to live with herself? The practicing linguist has been trained to conduct language description and analysis without oughttas, haftas or s’postas. What often happens with language descriptions is that linguists do their descriptive jobs to the best of their abilities, but through the process of language description and publication, inferences about what is or is not valuable regarding a language can affect what is done with those descriptive materials.
3.1 A false bifurcation
In this consideration of the process of creating a language description, the descriptive–prescriptive bifurcation has been shown to be inaccurate, impossible or unhelpful in various ways (see also Cameron, 1995; Milroy, 1992; among others). More options are needed. At the very least, the bifurcation needs to be expanded.4 The concepts remain useful, particularly if field linguists are aware of the concepts and how to appropriately use them during the field process.
There are multiple origins for prescriptivism. Some prescriptivism arises out of the necessities of accessing native-speaker expertise, including disagreements about the language. Some prescriptivism arises out of differences in priorities, such as the difference between a linguist’s idea that each language is inherently valuable and some language speakers’ ideas that being able to participate in the perceived prosperity of the majority culture is more important than retaining a language. Some prescriptivism makes language description possible but then arises on the other side of the description, such as in privileging a majority over a minority language form when creating educational or new language materials. And some prescriptivism can be used, depending on the people behind its uses, to protect or discriminate against speakers of minority languages. When even contradictory outcomes are possible, it is difficult for descriptive linguists to know how to create the most relevant and least damaging materials.
3.2 Where to go from here?
With several types of both prescriptivism and descriptivism available, the continued use of the bifurcation remains questionable. It might be clearer to those conducting research and those contributing to research and those using research if different terms were available for the different concepts required by the realities of field research. Discussions of variation, deviance (Chomsky, 1995; Zahedi, 2007) and idiolect are crucial in the development and applications of more specific terms that are predicated on the sources of prescriptivism and whether that prescriptivism is used to benefit or to harm. The concepts might vary depending on what arises in any given field situation, as well. If so, a reconsideration of the terms and concepts based on what is found in field situations could become necessary. Descriptive research might again become the driving force behind not just descriptive theoretical development, but also developments in explanatory theory.
Acknowledgements
