Featured image is Lower-Austrian Peasant Wedding, by Ferdinand Georg Waldmüller

J. L. Austin made a tremendous breakthrough in linguistics and the philosophy of language when he demonstrated the performative character of language—that is, by saying something we are always doing something. Extreme cases include “I now pronounce you man and wife,” which, when uttered in the right circumstances, changes the status of two people from being single to being married.

The problem with operationalizing this comes in with the notion of “in the right circumstances.” Can these be specified in advance? At what level of detail? How small do deviations need to be before the speech act is nullified (or “infelicitous” as Austin put it)? Are some infelicities more important or decisive than others, and does this vary for each sort of speech act?

Austin ultimately gave up on a completed system, though many speech act theorists since him have taken up the torch. Among these, his former student John Searle is the most notable.

But I stand with critics like Jacques Derrida and Stanley Fish in thinking that a high degree of uncertainty is required by the subject matter. Derrida has pointed out that if the successful performance of a speech act is determined by context, and context is boundless, then we can never know with the certainty of mathematical or logical necessity that we have avoided infelicity. There may be aspects of the speech situation that we did not notice at the time which invalidate it retroactively, and the uncertainty around this is ineradicable.

It is akin to digital security—we may use top of the line cryptography, we may use stricter than best practice implementations, but we cannot know about security holes that haven’t been discovered yet. If we could, then we would have discovered them already. There is no reducing, much less eradicating, uncertainty of this sort—in security or in speech acts.

The point is not that no speech act ever succeeds, but that it isn’t something we can really measure externally from the situation and the people involved. Moreover, even to participants it is not known with the certainty of the solution to mathematical problems.

Without pretending to such certainty, I’d like to build off of our previous discussion of Aristotle’s notions of actuality and potentiality, as well as efficient and finals causes, in order to continue the discussion of when speech acts go right or wrong.

We are not born in a vacuum; we begin our lives in a particular location within ongoing social games. You are not just a son or a daughter, but a son or daughter of your parents, who have certain positions and status in the various games they play. This has implications for you.

For one thing, it will determine what language you speak and later, when you are old enough to have peers, the way in which you speak it. Language use begins as a game children play with parents; parents trying to get their baby to name someone or something, babies trying to get a big reaction from parents.

As I said in my last post, some of our lower level potentialities are shaped by these games. To review, the ability to speak English is a lower level potentiality relative to the potentiality to speak any language. The games we play in infancy begin the process of developing our lower level potentiality for speaking and understanding a specific language.

I think that the structure of what Austin referred to as the total speech situation is created by these games, and the mechanism is shaping the potentialities of a large number of players to point towards some specific actuality in concert. Changing the status of a couple from unmarried to married becomes the final cause which requires some specific conditions to act as efficient cause.

The marriage license and wedding ceremony, administered by someone with the appropriate authority, with witnesses, act jointly as efficient cause. But this cause works differently in each individual. To simplify, let us say that people have the potentiality to treat other people as married, which has implications for how certain social games are played with them. People also have the potentiality to be married to someone, which has some other specific implications for how they play social games with their spouse and with people outside of their household.

Coming into the wedding ceremonies with these potentialities shaped by similar social games, a single speech act is capable of actualizing—that is, serving as the efficient cause for—these different potentialities to create something akin to a unified situation, in which the couple are married in the eyes of all (including themselves).

The unity is only relative, however. It can only be preserved so long as others who are or will become connected to the couple are successfully invited to play the relevant social games in a particular way. For these people, who were not at the wedding, it may be the recognition of the couple’s peers, or the government, or simply the couple’s own word that they are married, which accomplishes this.

But it is not hard to imagine situations in which this goes awry. If one of the spouses becomes a pariah in the community, or if each spouse is of a different background (racial, religious, ethnic, regional, take your pick) and the community suddenly begins to reject such unions, they may suddenly stop being recognized as married. The laws do not even need to change, if local law enforcement, and judges, and juries begin to act as if they had. And people with authority on matters of law need not be the ones to act differently in order for the couple’s status within the community to have changed.

It is perhaps easier, and less horrifying, to think of this in terms of money. People can stop recognizing a given currency as money at any time. Even when a government facing runaway inflation manages to strong-arm its population into continuing on with the official currency, it is clear that losing people’s willing recognition changes the status of that currency in important ways (not the least of which is to contribute to the inflation).

Humans clearly have some higher level potentiality to treat something as a medium of exchange with general purchasing power, though how high a level this is—that is, how much at even this stage it’s largely shaped by social games—is a question I’m not in any position to answer. Nevertheless, the potentiality to treat some specific item as such, and to recognize it as a dollar or a peso or what have you, is definitely shaped by social games. Being a person who has a lower level potentiality to treat dollars as currency is the final cause in a situation where what people in our lives have said and done acted as the efficient cause. Once we get to that point, simply telling us the price of the thing we wish to purchase is enough to get us treating dollars as currency in actuality, in the moment.

Now, the language-games of Ludwig Wittgenstein, and the hermeneutic games of Hans-Georg Gadamer, have a much more fluid notion of “rules” than social scientists are used to. The latter like to think of rules as determining behavior, or at minimum forming the boundaries. For the former, the structured behavior is constitutive of the rules, rather than determined by them. We should see the rules of a game as something we are invited into, and certainly something that is very influential, akin to the rules of grammar. But just like the rules of grammar, they can be changed by our own actions. And they are not changed the way that legislative rules are changed—we do not draw up a new system of grammar and implement it. If enough people start to behave according to different rules, then the rules have simply changed. It is a matter of enacted practice, rather than higher level articulation.

The potentiality for these structural shifts—in language, currency, and social relations in general—must already be in place for them to occur. The potentiality for this type of change per se must be fairly high level, but the potentiality for specific changes is lower level. The fact that a new word can suddenly be widely adopted, or an old one used in a specific new way, suggests that a lot of people had similar pre-conditions which allowed that to come about. At a certain point, of course, the efficient cause of more people adopting the new word or word usage is simply that so many others have already done so.

This, then, is my tentative answer to Austin and Derrida: the pre-condition to a successful speech act is that social games have aligned the potentialities of the participants in the speech situation, so that the speech act is able to act as an efficient cause, actualizing the specific potentialities which create a new situation. But the unity of this new situation is always relative, and the success always partial, because the persistence of the change depends upon inviting people outside the situation to play social games in a particular way moving forward. Current participants may also change how they play in the future as well.

So while I think it is possible for us to sketch out the basic outline of how this works, as I have done above, we still cannot do away with the irreducible uncertainty inherent to it.