Motto:

"There are none so blind as those who will not see." --

Sunday, August 20, 2006

On Stipulations

I want to consider a few questions here: What are stipulations? Also, what is their function? What exactly is it that they (can hope to) accomplish? In what sense can a stipulation establish synonomy?

It seems there are a few basic candidates for what a stipulation could be:

1. A report, on the part of the stipulator, of their use of the term being defined/introduced.

2. A perscription/reccomendation that the stipulator's audience use the terms flanking the equivalence sign in the same way.

3. A resolution on the part of the stipulator to use the terms/expressions flanking the equivalence sign in the same way.

4. The truthmaker for the claim that the expressions flanking the equivalence sign have the same meaning.

...and so on. Perhaps stipulations have only some of these traits, perhaps only one such trait. Perhaps they have all. Perhaps some have some while others have others, giving the category a kind of "Family Resemblence" structure. I think the cheif puzzle here is how to account for the seemingly magical ability of stipulations to bring about certain semantic facts, the synonomy of different terms, by sheer fiat. After all, I cannot stipulate that water runs uphill and seriously expect it to be so--how is it any more possible to stipulate that a given expression means that water runs uphill? The problem seems to be that such semantic facts would be insulated from usage. According to (1), stipulations will be true or false depending on whether or not the report of the speaker's usage is accurate or not. In this case the status of the stipulation is will indeed be posterior to usage, but since it is merely a report of the speakers usage it will not be able to do the synonomy-establishing work stipulations are normally taken to do . On (2) and (3) stipulations will lack truth value. Yet it is not clear what stipulations, thus viewed, are supposed to accomplish--for perscriptions might not be obeyed, and the speaker might not follow through with their resolution. The terms will only be synonomous for those who use them synonomously. Stipulations would only bring about synonomy in a causal sense. On (4), saying that two terms are synonomous will amount to nothing more than that a certain speech act took place. For if the speech act itself is the truthmaker for the claim that the terms are synonomous, they will remain so despite any subsequent divergence in usage. An odd consequence, if you ask me.

2 comments:

Anonymous said...

I think you have oversimplified the situation a bit. On 2 and 3 particularly, you might want to consult the latest literature on non-cognitivist ethics; making stipulations lack truth value may not leave them as impotent as you suppose.

Perhaps pressing on point 2, though perhaps this is a hybrid of points 1 and 2, I would also point to the work of Lewis on conventions (as well as the more recent view of Millikan on functions, which appears to me to be closely connected). What makes a practice a convention is that it is perpetuated as a result of a purpose that it serves, where the fulfilling of that purpose requires a general expectation of at least some faithfulness to the convention.

Since in order for a convention to exist, it must play this role, it is not a mere report of usage; very consistent usage may not constitute convention, and in some cases even very inconsistent usage may still be consistent with the presence of the convention. The intention to establish a particular usage indeed probably suffices for the establishment of a convention in most cases, though if the prescription comes to be forgotten or completely ignored the convention will fade out of existence.

One might question whether the mere intention and issuance of a prescription is sufficient; what if the forgetting and ignoring follow immediately? In that case, I think that if this is a deviant occurrence, we may wish to say that the convention simply faded very quickly; the convention was established by the intention and immediately vanished when the intention vanished from memory. I would also further say that a circumstance where such forgetting and/or ignoring always occurred would be a situation where I'd have some doubt that intention formation or prescription were possible.

I suppose I would conclude by noting that it would seem to me to be even odder if it turned out that we could not introduce new technical terms by stipulation, since it seems to happen so often.

Unknown said...

Thanks for the recommendations on Lewis and Millikan, I probably should check them out.

I think I didn't express myself as clearly I could have. My point wasn't that new technical terms can't be introduced via stipulation, but that stipulations are not, shall we say, intrinsically efficacious. I do think they can bring about synonymy, but I think they do so mediately, not immediately. That is why I said "The terms will only be synonomous for those who use them synonomously. Stipulations would only bring about synonomy in a causal sense." On my view, then, stipulations as understood on (2) or (3) are without effect unless the prescription is actually carried out by the stipulator or their audience or both. Thus I think that if both the stipulator and/or their audience fail to carry out the prescription, the stipulation by itself is not enough to bring about the synonymy of the terms. In principle stipulations could be dispensed with; their work could be carried out, e.g., by the author politely asking their audience to use the two terms in the same way.

Given the above, I’m a little puzzled by what you say about conventions. For example, you say “The intention to establish a particular usage indeed probably suffices for the establishment of a convention in most cases…” and later on: “One might question whether the mere intention and issuance of a prescription is sufficient; … we may wish to say that the convention simply faded very quickly; the convention was established by the intention and immediately vanished when the intention vanished from memory.” I might be misinterpreting you, but to me this suggests that conventions can exist, if only briefly, without anyone practicing them. If they cannot, how could the intention be sufficient to establish them? Yet if they can exist (briefly) without the intention being carried out, I have to wonder just what sort of things conventions are supposed to be. To clarify, let’s suppose we have a case of a new technical term being introduced, by stipulation, as being synonymous with a more familiar but also more cumbersome expression. If the presence of a convention is compatible with inconsistent usage, does that imply the two expressions could have the same meaning despite a subsequent wide divergence in usage? Or does it imply instead that the presence of a convention does not require sameness of meaning?