Shifting Languages: Interaction and Identity in Javanese Indonesian. By J. Joseph Errington. Cambridge: Cambridge University Press, 1999. 216 p., bibliography, index. Reviewed by Rita Smith Kipp (Kenyon College).

The word “shifting” in the book’s title alludes both to the macro process of language shift (when the knowledge and use of two languages in an entire community changes) and to the micro, conversational process of code-switching. This book, based on research during 1985-86 in Solo and two outlying mountain hamlets, examines some excerpted conversations that Joe Errington’s research helpers recorded on tape when he was not present. The book begins with the question of whether Indonesian is displacing Javanese, and while answering no to that question rather quickly, it ends with the more complicated prospect that these languages are no longer “two distinct autonomous systems”(186).

Does Javanese usage signal identification with an ethnic in-group? Javanese is bifurcated into a low (ngoko) style associated with intimacy and informality, and a high (basa) style that indexes hierarchy. Everyone commands ngoko, but only elites command basa. Some elite Javanese, engaged in discourse about preserving tradition, worry that basa is losing relevance for the young. Fluency in it is increasingly a commodity at events that are supposed to unfold according to “tradition,” or where being Javanese merits demarcation. Expert speakers, termed protokol, hire themselves out as masters of ceremony and use basa monologically to address the assembled audience.

If Javanese is a native language, Indonesian is an un-native language (as opposed to a non-native language), in that it is spoken here only by those for whom it is a second tongue. It is thus “an out-group language without an out-group”. (3) Although Indonesian is a “detached, ‘third-person’ way of speaking,” (97) speakers use it less to signal otherness than strategically to appropriate its associations with modernity, rationality, and the bureaucratic authority of the nation state. Sometimes people narrate something in Javanese, then draw their conclusion or an abstraction using Indonesian.

Speakers usually use Indonesian, furthermore, thoroughly mixed with Javanese (and vise versa) in a “language salad” (bahasa gadho gadho). One of the chapters illuminates so-called discourse particles, those ubiquitous, enigmatic emphasizers such as lho, kok, lha, and ta that Javanese sprinkle into their salad sentences. Some lexical ingredients are unambiguously Javanese or Indonesian, but some are hybrid forms, e.g., words made from Indonesian roots with Javanese affixes. The result is, as one of Errington’s assistants commented about a particular recording, “There’s a lot of Indonesian but it feels very Javanese” (113).

One of the most interesting chapters has to do with the phenomenon called speech modeling. This is when a speaker suddenly dispenses with “he said/ I said” when recounting a conversation, and instead, momentarily reenacts the dialog, speaking all the relevant voices as if in verbatim recall. This moves the addressee from the implicit position of the second person into that of bystander or audience. These modelings sometimes evince the empathy of the speaker with those he or she is describing (or addressing -- since speakers sometimes model what they suppose their listener might have thought or said). There is even a term for modeling one’s own speech or thought -- ngunandika.

Why, when speaking Javanese, do people insert these little performances into their conversations? Errington was frustrated by the fact that neither speakers nor his assistants wanted to impute any strategies for this behavior to others or even to themselves, unmoved by his persistent questions about this opaque but (to them) wholly irrelevant issue. (Clifford Geertz once summarized fieldwork as, “plaguing subtle people with obtuse questions.” Three of Errington’s helpers quit, so bored were they by the tedium of transcription.) Errington could make sense of specific instances of speech modeling after the fact by locating each within a conversational context, and by drawing on ethnographic literature by Ward Keeler, Jim Siegel and others. Perhaps speech modeling is another instance of “authority ventrioloquated in power-laden dissimulations of self” (137). While Errington can recover the impetus for such modelings after the fact, he is not likely to develop rules to predict or describe this kind of linguistic behavior.

Errington compares the slippage between narration and acting to another kind of shift ­ that between ngoko and basa. One recording captured a speaker relating a story to two interlocutors, one of whom, a woman near the speaker’s age, would have normally elicited ngoko from her, and the other, an older male with whom basa was more appropriate. Her narration unfolded mostly in basa, but veered into ngoko when directed more toward the woman. It also included some self modeling (ngunandika) entirely in ngoko .

Anyone who has read much linguistics would know not to expect an easy read here. The prose seldom flows when one is writing about talking. But the materials are rich, descriptively thick, and fascinating. One annoying technical flaw throughout the book was the odd insertion of hyphens in words written mid-line, errors that should have been caught in proof-reading. There is much here, however, to repay our patience with these detractions.

(This review was uploaded to www.antarakita.net on September 26, 2003)