I’ve written a number of posts about “the ask” and why we should insist that ask is a verb, but I haven’t said much about the provenance of the ungainly nominative “ask” or taken its origins into account.

“The ask” is not the revival or survival of an archaic or medieval form, as one writer in the New York Times suggested. Far from it: it’s a piece of stock trader’s jargon that crept from the trading floor into bureaucratic conversation. On Wall Street, “the ask” is shorthand for the minimum price a seller sets for a security. The difference between the ask and the bid, or what a buyer is willing to pay, is known as the spread; and the spread is one pretty reliable measure of market liquidity. 

Presumably, when someone uses the term “the ask” or “my ask” to direct work or coordinate action, he expects (or pretends to expect) the second person, his interlocutor, to counter with a bid, as if conversations produced a workable measure of practical liquidity — or a measure of what’s practically possible — in the difference between what one person wants to do and what another wants or is willing to do, or what each thinks ought to be done.

We can sketch a model: the ask would invite the bid and the bid would meet that invitation with an offer. And we can develop this rudimentary model of conversation a little more by exploring the etymology of the word “bid” — itself the substantive form of a verb with roots in Old Teutonic, where *beudan means to stretch out, reach out, offer or present; and by extension beodan or boden in Old English and bede in Middle English come to mean to announce, proclaim or command. So here, again, an ask-bid model might confer some power on the bidder, or help create the appearance of parity, a sharing of command between petitioner and respondent, asker and bidder. What we are going to do is what we together command, or what will fall within the spread, span or scope of our shared command.

That doesn’t seem so bad, on the face of it: at the very least it sounds as if people on both sides can give and get in return. “The ask” holds out the promise of some share in power, or at least more flexibility than command-obedience would seem to allow. That might help account for its widespread use in the first decade or so of the twenty-first century. Ideas about organizational hierarchy are changing, and people have begun to pay outward homage, at least, to the idea that command and control is not necessarily the most effective way to run an organization. In bureaucratic settings, the imperative of command is taking on interrogative affects: the ask makes an order sound more like a request, softening the power one person actually wields over others.

The model has lots of shortcomings: for one, it reduces human relationships to market transactions — and that’s a serious and thorny problem, one I hope to say more about in a future post. But the main trouble with the ask-bid model is simply that it tells us very little about how conversation actually works. Conversations are never so neatly regimented and sequenced as this bureaucratic model makes them out to be, and as I wrote in another post, much of which we might regard as background noise or “beside the point” in a conversation is just as important, if not more important, than the putative point. There’s never just “an ask”; all parties to the conversation are continuously asking and offering, requesting clarification or confirmation, making representations of the other, shifting attention to and from the matter of joint interest, situating, interrupting and re-connecting with each other.

Generally, we’re making it up as we go along, together, and all of that joint effort counts much more than we ordinarily acknowledge. We don’t merely counter asks with bids or requests with offers; we also work together to organize, represent and sustain the conversation as a social act.

A paper published last week by Mark Dingemanse and others at the Max Planck Institute for Psycholinguistics stresses this theme. In ordinary conversation, the authors observe, interlocutors ask for clarification and correction about once every 1.4 minutes. This “other-initiated repair” is a central feature of all human conversation; and the cues one uses to initiate repair demonstrate care for the interlocutor and for the “social unit” of the conversation.

There are three main ways interlocutors initiate repair. Interjections like “huh?” are “open requests” for clarification. Asking speakers to repeat what they said (“who”?) are “restricted requests.” Repeating back what the speaker said (“he hit a homerun?”) are described as “restricted offers.” All three are regularly used with the same frequency no matter what language we happen to be speaking and despite differences in grammar and syntax. (For the paper’s authors, this lends support to the hypothesis that there are universals at work in all human language; but rather than get hung up on that point, I prefer simply to appreciate their observation that interlocutors are working constantly together, making repairs on the fly.)

So instead of a simple ask-bid or request-offer model, we need a much looser and less linear model in which all parties to a conversation are constantly running requests and offers and making interjections in no particular order or sequence, and so frequently and effortlessly that we don’t even notice we are making them. Requests, offers and interjections might even go unanswered; but they are no less effective for all that. These are cooperative cues and gestures, markers of the conversation as a social act.

(This is, by the way, why,so-called “conversational” interfaces built for digital assistants like Siri are still nowhere near conversational. What the manufacturers of these devices really mean is that you can address your digital assistant — give an order or make a request — in ordinary language, and it will follow. But even if the assistant is designed to say “I did not understand your request, please repeat it” or something along those lines, it’s not producing anything like the steady stream of other-initiated repairs involved actual conversation, where interlocutors are reading each other’s minds and correcting misreadings as they go.)

Conversation recreates and demonstrates joint commitment. That’s what’s really missing from the ask-bid model: human relationship. Asker and bidder, seller and buyer, don’t have a shared project beyond the exchange they are negotiating; their contact with each other can end once the transaction is made, and one or both can just walk away if they don’t agree on price. After all, both the asker and the bidder seek advantage over the other, rather than mutual gain or shared advantage that is the spur, aim and outcome of serious conversation. After the deal is done or abandoned, the bidder is free to pursue his ends and the seller is, too, even if they will be working at cross purposes.

On the other hand, people who are in a conversation about what to do have already committed to doing something together. They’ve committed to acting together, to social action and to a social subject: a “we.” We keep our commitment by repairing as we go. We act together even when we have irreconcilable differences about the way things are or what to do.