In which I interview an article: Silverstein on indexical order

Q. With us in the studio this evening is Michael Silverstein’s 2003 article “Indexical order and the dialectics of sociolinguistic life,” published in the journal Language and Communication, vol. 23. Now, you have a reputation as being, shall we say, a little bit hard to get to know. I’m going to play back a quote from you:

But where does the authoritativeness of the logic of valorized partitions of social space come from? What convinces sincere believers-in-essences that such essential characteristics are “truly” predicable of things? It is, of course, the nature of what we recognize to be ritual, or at least relatively ritualized tropic invocation of essentializations (naturalizations) to make believers of us all. Ritualization’s own micro-contextual semiotic modality is indexical iconicity, by which a ritual(ized) text appears to achieve self-grounding in the (relatively) cosmic absolute of value-conferring essences. Every macro-social framework in which micro-contextual indexicality is locatable seems to be centered on certain relatively ritualized manifestations of the indexical signs in organized configurations that license or warrant their occurrence elsewhere by a kind of either historically-unique or legitimately-recurrent stipulative or “baptismal” essentialization of their power as indices linked to a larger macro-sociological system. (p. 203; emphasis in original)

When I read that section, here’s what I wrote in my notebook: This seems important but I don’t know what it means.

A. Well, clearly what it means is —

Q. Let’s try and take this apart starting at the beginning. It seems to me the main question you’re dealing with is, “Why do words mean what they do?” But when you say “meaning,” you’re thinking in a broader sense than how that word is typically used.

A. Right, one thing you know when I say a word, I’m talking about a thing in the world. Like I could say “wine” to refer to that fine bottle of wine over there. You know what “wine” means, obviously. But I could also call it a “Semillon / Sauvignon-Blanc blend,” or “the Château Haut-Brion,” and then I’m telling you more about the wine itself, but I’m also telling you more about me, as a person who knows and understands these things about wine, and feels it’s important to point them out to you. That’s really part of the meaning of what I said. I’d say, the more interesting part.

And it’s not just words either. For example, think about when you’re learning French and you have to figure out when you say “tu” and when you say “vous” (cf. Brown & Gilman 1960), you’re communicating something about yourself by your choice there as well. Well OK, that’s also words, but in Javanese, you know, there’s a six-way distinction instead of a two-way distinction to indicate degree of deference, and it involves different grammatical forms in different parts of the sentence. Or to give another example, if you’re from New York and you pronounce the word “source” so it sounds like “sauce” (cf. Labov 1972), that also has a social meaning, just in the pronunciation. Or if you think about different speech registers? Like, that quote I just gave you a moment ago, you can tell that it was written by an anthropologist — not just because of the technical jargon like “indexical,” but also the sentence structure, and the way I use rhetorical questions and adverbs like “relatively.” You expect those features to all come together, and when you see them all together, you think “anthropologist.”

Q. Oh, so that’s what you mean by “enregisterment.”

A. Right.

Q. And your claim here is that all these different examples like tu/vous, source/sauce, wine/Sauvignon-Blanc, and this specialized language of anthropology … that all of these examples are really specific cases of one more general phenomenon. Including grammar, vocabulary, pronunciation, discourse, whatever.

A. Right, all those labels are not so interesting to me. I think looking at it as language structure, as you’re doing by using words like “grammar” and “pronunciation,” blinds you to what’s really going on here.

Q. Which is what? Familiar versus formal, something like that?

A. Not a bit. You’re talking like one of those “‘billiard-ball’ sociologists” (p. 197) who treat social context as an objective thing, as if you could come in and identify “formal” contexts empirically, whatever that means, and then measure the effect on people’s language use quantitatively. When of course we know that context is constructed “in-and-by social interaction” (passim). It’s a dialectic.

Q. Okay then, so if it’s not familiar versus formal, then what is it?

A. Indexicality.

Q. …

A. Okay, so when you say a word — and when I say “word,” it’s a simplification, I really mean “semiotic unit” — when you say a word, you choose one that’s appropriate to the context. Your word choice presupposes something about what was going on before. At the same time, that word also keeps the ball rolling, from an enregisterment point of view. It’s effective in the context, as it’s ongoing. It entails a certain way of understanding, and it entails that other people will respond to you in a certain way.

But the tricky thing is, how do we know what the context is? I mean, not as analysts, but as social actors, as people talking to one another. What are the unspoken rules that are governing the interaction between us, and how do we know to follow them?

Q. This is “metapragmatics”?

A. Exactly. Sometimes we do this explicitly, like when you called me up and said “I’d like to interview you,” I knew it was going to be an interview, and I would treat it as one, speak in appropriate interviewese. But more often, and more interestingly, it’s implicit. You can infer the metapragmatic context through “token cooccurrence patterns of emergent entextualization” (p. 196).

Q. Say whaa? No hang on, let me figure this out. Token, you mean as in a type/token distinction. Every word — sorry, “semiotic unit” — that you say is a token of a type. There’s the act of me saying “wine” at one specific time, and then there’s the abstract notion of the word “wine.” So, tokens, we’re dealing with words as they’re used by individuals in specific contexts. And “cooccurrence patterns,” that’s what we were saying before about enregisterment. The meaning isn’t so much in one word as it is in the overall effect of different words being used together that are associated in some way. And then “emergent entextualization,” we’re not thinking about large-scale registers, but about the way that each example of text or social interaction unfolds. And then getting back to metapragmatics, that process is how we communicate our understanding of what kind of social context we’re in. Am I right?

A. More or less.

Q. So basically, you mean discourse genres? When we recognize the genre, we know how to participate in it? Or, like, A. L. Becker’s (1979) definition of “context,” involving cultural constraints on how texts are related to one another, speakers, hearers, nontextual events?

A. Now look, I meant it the way I said it.

Q. Right, of course.

A. But then, if we have this idea of metapragmatics, it basically shows the way to answering every question we have about language in society. It’s not statistical, and it’s not about cognitive rules or anything like grammaticality. It’s about making hypotheses of what’s going on, from moment to moment, in real time. The only way we can understand one another is that we each have in mind a model for “this kind of interaction,” however we’re classifying it, and we interpret everything through that lens.

Q. So, indexicality.

A. Right, I’m getting there. Now remember that people and events don’t just spring up out of nowhere. We each have our own culture, beliefs and ideologies that explain how the world works. One function of that is what I was just saying, about “kinds of interaction” — we have all of these individual communicative experiences which we then project onto a macro-level and say “these are all tokens of the same type,” essentially, “these are all the same kind of situation.” And then we develop beliefs about that kind of situation. It becomes essentialized.

So, indexicality — let’s think back to that long quote you reproduced up top. The words index one thing and another, the grammatical structures index one thing and another, and eventually that builds up into an n-th order indexicality that tells you “This is how anthropologists write.” But then, how do you feel about anthropologists? I’m sure you find them insightful and erudite, not to mention more attractive than the average social scientist, but someone else might find them needlessly technical and obscurantist. That’s your ideology, as regards anthropologists. And that becomes the n + 1st order indexical meaning of the anthropological register. It’s different for different people, because they have different ideologies.

So getting back to your initial question, I don’t like to say that words “mean” anything in particular, because the words aren’t doing anything. They’re just tools that we use to communicate. Rather than meaning, they index all sorts of things. And those indexicalities get layered on top of one another, they’re sites of social conflict and loci of diachronic social change. The social categories themselves are changing over time. For example, some people would say that if you pronounce “source” and “sauce” alike, within the context of New York, the basic meaning of that is lower socioeconomic class. But that presupposes that socioeconomic class, as we understand it, is some kind of essential category rather than something that is itself socially constructed. That’s why I’d call it n-th order. And then we can consider where it comes from, what underlies it, what can be built at higher orders on top of it.

Q. Well thanks, it’s been a real pleasure having you.

A. Thank you. Great to be here.


Becker, A. L. 1979. Text-building, epistemology, and aesthetics in Javanese shadow theater. In The imagination of reality, ed. by A. L. Becker & A. A. Yengoyan. Norwood, NJ: Ablex.
Brown, R., & A. Gilman. 1960. The pronouns of power and solidarity. In Style and language, ed. by T. A. Sebeok, 253–276. Cambridge, MA: MIT Press.
Labov, W. 1972. Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press.

Tagged with: , , , , ,
Posted in Uncategorized
8 comments on “In which I interview an article: Silverstein on indexical order
  1. Believe it or not, this just solved a problem I’ve been having about a paper I’m presenting at the Society for the Study of Midwestern Literature. Thanks! Will have to read Silverstein.

  2. […] recently read another entertaining post in which a grad student in linguistics, in an attempt to come to grips with a difficult academic […]

  3. Martin Kurani says:

    Really, really clever and helpful! Thanks!!!

  4. susanblum says:

    I gave this clever piece to my graduate students who found the article….challenging. Thanks for posting it!

  5. I’ve just been assigned this reading and I can’t make heads nor tails of Silverstein’s rather opaque writing style. Thus, I was both amused and relieved to read your comment, ‘This seems important but I don’t know what it means.’ His comments somewhat helped to illuminate his theory. Thanks for posting.

    • Daniel says:

      Thanks, glad it helped! To be clear, they’re not his comments—the conceit is that I’m interviewing the article itself. Michael Silverstein was not involved in the production of this blog post.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s