The issue is what the truth maker is of a certain type (subset), L, of propositions. A proposition belongs to L iff it has the form “X is the correct way to spell the word.”.
Truth makers and truth guides
A truth maker is what makes something true. ‘makes’ here is a semantic relation. What makes the proposition ‘it is raining outside’ true is the fact that it is raining outside. Another way to put it is to ask under which conditions the proposition ‘It is raining outside’ is true. And the answer is exactly when it is raining outside. Such a condition is called the truth condition.
A truth guide is a way to discover what the truth is. Asking your math teacher what the solution to an equation is, is probably a good way to discover the truth. However it is not that the teacher thinks the answer is x that makes it true that the answer is x. And if the teacher is not infallible, and no teachers are, then he might be wrong. It may be hard to distinguish between an infallible truth guide and the truth maker but we need not concern ourselves with this since it is never the case that a truth guide in infallible.
Language claims about spelling
I’m trying to discover what the truth maker is of L-type propositions. So far a number of possibilities have been suggested to me:
- How the majority of the general population uses the word.
- What the majority of the general population thinks is the correct way to spell the word.
- What some special dictionary says is the correct way to spell it.
- What the majority of dictionaries say is the correct way to spell it.
- How the majority of some subset of the general population uses the word.
- How the majority of fluent, educated speakers use the word.
- How the majority of fluent, educated speakers and serious writers use the word.
- How the majority of serious writers use the word.
The list goes on but I believe the above captures some of the good initial guesses. What makes it so hard to determine is, I think, that even though there is only one truth maker all the possibilities function as truth guides.
Analyzing (1) and (2)
(1) seemed like a good proposal to me, that is, for all I knew it might be true. However when analyzed closer it becomes clear that (1) is not a good proposal at all. All we have to do to see this is to ask ourselves what ‘general population’ we’re talking about. Suppose we’re talking about the danish language. The general population seems to be danes. But then how about other people that can speak danish? Danish is taught in schools in other countries like Greenland. And how about danes that can’t speak danish? Presumably some people have managed to become citizens of Denmark without being able to speak danish. Do they belong to the general population? It seems not. The general population then is not a good idea at all because it is unclear what this refers to, and if it refers to speakers of danish, then it is equivalent to (5).
Similarly (2) becomes an extended version of some (5)-type proposal.
Analyzing (3) and (4)
(3) also seemed a good proposal to me but again when analyzed closer it loses its status. When I proposed the theory Pyrrho wrote:
“You also have the oddity of there being, according to your theory, no right way to speak or write before dictionaries were invented, which is very curious indeed. Was there not a difference between fluent speakers of English and those who were not before the invention of dictionaries? Was there not a difference between being very literate and not so literate before the invention of dictionaries?”
That is a curious implication indeed. One way to fix it is to say that the truth maker of L-type propositions changed some time after dictionaries were invented. This is obviously an ad hoc assertion and makes the theory less simple.
Another oddity arises for (3), as Pyrrho writes:
“Yet another oddity of saying a particular dictionary is the “truth maker” is that it would mean that there could be no such thing as a misprint, as it being in the dictionary, according to that theory, is what makes it true. It is a very odd theory that postulates the impossibility of a misprint.”
Indeed that is a very odd implication. Clearly it is not impossible to misprint and thus (3) is false.
One obvious problem is to discover what the necessary and sufficient conditions are for a given person to belong to the group of people’s who makes L-type propositions true.
A ‘Fluent’ speaker means a speaker who is “able to speak a language accurately and confidently.” It is unclear why fluent speakers have anything to do with L-type propositions since they do not necessarily write anything. Most people do not write much but speak a lot and most speak fluently too. Why does it matter how they spell a particular word when they rarely write it?
It is unclear what ‘educated’ refers to in this case. Educated about what? The language? Are we talking about a formal education or will an informal doo? Most of one’s first language is probably not learned via any formal education but by mimicking parents and other people around one. ‘Educated’ is too vague to be a useful condition.
It is also unclear what a ‘serious writer’ is. Perhaps it’s just someone who writes a lot. It would be odd if only how a minority of writers wrote was the truth maker. How does one decide if the writer is serious in that sense? It seems that disposing of the unclear ‘serious’ is a good idea and just settle on ‘mere’ writers of the language.
Another trouble is how to deal with the majority thing. Suppose that there are n writers that write a particular word in one way and there are n writers that write the same word another way. Suppose further that n+n is equal to the whole population of writers. Which way is the correct way? There seems to be no answer to that question. Indeed the best idea seems to be to reject the question and say that there are two correct ways. Suppose we accept that. Again it can be asked what the correct way to write a particular word is if the group of writers is exactly divided into three groups. Again one might answer that there are three correct ways. However if we continue this way we will reach the conclusion that if all writers wrote it in their own way, they would all be correct. This is a curious implication.
However in practice this is not a problem since it is never the case that writers are that divided. In reality is suffices to say that if writers are divided somewhat equal into, say, five groups, then all these are acceptable and that they ought to find a particular way to spell the word to avoid confusion.
One thing remains to discuss about (5) and that is whether it is how writers use the word or how they think it is spelled correctly that is the truth maker. I know of no way to decide this question since they correlate so much: Writers write the way they think is correct and writers think the way they write is correct. It doesn’t seem like an important question to settle this.
The majority of dictionaries theory (4) and the majority of writers theory (5d) seem to be equal in explanatory power but since the writer theory is more simple, then we ought to prefer it.