Few philosophical methods exist in a vacuum. Anyone who endorses or uses a philosophical method thinks that it will serve the purpose that it is intended to. Often that purpose is to produce true theories, but I won’t presuppose that. In any case, the endorsement of the method usually rests on some other philosophical position, which implies that the method will achieve its goals, that the theories produced as a result of it will meet the standards that they are supposed to. For example, one philosophical method might endorse certain kinds of intuition as true, and from those intuitions construct philosophical theories. This method itself is then justified by an epistemological theory, which claims that these kinds of intuitions are true, or at least likely to be true. The epistemological theory is then the foundation of the philosophical method itself, because it justifies the method.
Today I am going to consider philosophical theories about meaning as foundations for philosophical methods, and what standards methods endorsed by them can or can’t meet. But first, why would anyone allow their philosophical method to rest on a theory about meaning? The meaning of words usually seems to be, at best, peripheral to a philosophical investigation. When we consider the political structure of a society we are not really that interested in what exactly the word “democracy” means, democracy is just a label we use as part of our theory. To understand how theories about meaning ended up in a foundational role we must look to Aristotle. Aristotle thought that knowledge was to be had by grasping the essences of things, and that a proper definition captured the essence of what it defined. Putting these two ideas together we come away with the impression that is we could simply properly define things then we would grasp their essence and have knowledge about them. (Note that I do not think Aristotle himself proceeded in this way.) Suppose then that you have a theory about meaning that allows you to determine on the basis of observation what the meaning of the word is. Then you have grasped the essence, and so have knowledge about it upon which to base your philosophical theories.
Of course Aristotle is not the only reason that theories about meaning ended up in such a foundational role. Another aspect of philosophy responsible for this is the tradition of trying to answer philosophical questions in the form “what is X?”, where X is something of a debatable nature, such as justice. It seems clear that to answer such questions we must have a theory about what meaning is, in order to allow us to arrive at the desired definition. (Whether answering these questions is even a proper goal of philosophy is something that I will return to shortly.) The beginnings of analytic philosophy also emphasized language, and the idea that all philosophical problems were really linguistic confusions was popular. Thus it seemed reasonable to attempt to understand the meanings of words and then, on the basis of them, show how the apparent problems disappear. Of course it is possible for the meanings of words to imply these confusions, there is no guarantee that all words work well together, and so some later analytic philosophers have interpreted their task as correcting or improving upon the normal meaning of words so as to eliminate these confusions. And, in addition, the very description “analytic philosophy” implies that its task is to produce analytic truths, those true by the virtue of the meaning of words.
To say anything about the methods that rest on a theory about meaning I must make a few quick remarks about it. Actually only one quick remark, namely that words have no fixed meaning. The same word may mean different things at different times for different speakers. Thus such methods cannot produce universal “truths”. Now not every philosopher is attempting to produce true theories, but many are, so let me address them first. I will admit that, in a technical sense, claims based on the meaning of words may be true in the sense that they may truly follow from the meaning as understood by a particular speaker of the language at a particular time. But this does not seem to be what these philosophers were hoping to achieve. If it is claimed that “justice is a virtue” this is not a fact about justice, this only reflects that justice, as the word is used with one particular meaning, is a virtue, under one particular meaning of that word. But someone else could, with equal honesty, assert that “justice is not a virtue” if the meaning they give those words is different. The idea that we could have true statements in this form is based on the theory that justice exists “out there” on its own as a kind of object and that we can make true claims about this object and its relation to other abstract objects, just as we make true claims such as “the table is in the kitchen”. The claim that “the table is in the kitchen” spoken under a specific definition of the terms reflects a truth about the objects referred to. And it is desired that in this same way statements such as “justice is a virtue” to reflect truths about the abstract entities designated by them. Even if we grant that such abstract objects exist there is still a problem, these truths do not follow from the meaning of the words but what they happen to refer to. To make true statements in this way we have to be able to have access to the objects being referred to themselves, because or conceptions of them may be in error. And it is clear that because people can disagree over what words mean that we do not have the necessary access to the abstract objects, even if they did exist. Now perhaps this can be avoided by making the reference of words such as “justice” not an abstract object, but simply the actual things or situations people call just, and certainly we have access to those things (ordinary language philosophy). But then we run into a problem of inconsistency, because the theory about meaning refutes itself. What people say that meanings are, or the meaning of words are, are rarely the things they most often designate with them. And so meaning cannot be what the theory of meaning says it does. Thus the theory implodes. This is why I implied earlier that answering questions such as “what is X?” may not be the proper goal of philosophy, answering them can only be done in a conventional way, we cannot properly answer them in a true way.
Of course if we aren’t looking for true philosophical claims then the problems raised above aren’t really problems at all. But then other difficulties face the philosophical method. One problem, which will face any philosophical method that doesn’t intend to produce true theories, is that it seems to undermine the method itself. The method being a good method is predicated on the theory of meaning that justifies it being a true one, because if it is false the method will fail to achieve what it strives to. The problem here is that practitioners of the method are committed to the idea that philosophical theories aren’t the sorts of things that are true or false, and yet they are tacitly accepting a particular theory of meaning, a philosophical theory, as true. A second problem is that the meanings of words may be inherently contradictory. The meaning of many words comes in “layers”. For example, people may call the laws of their country just, but they also expect justice to be good. But the laws of their country may fail to be good on occasion, and so their understanding of justice contradicts itself. A theory based on the actual meaning of words will thus contain inconsistencies as a result of its sources, or will omit some aspects of the actual meaning of words, thus diverging from its purported basis in meaning. Finally, a philosophical method that purports to be unraveling the confusions created by language does not need to be actually employed. If we have reason to believe that all philosophical confusions are created by confused language the best approach is simply to set such problems aside and refuse to address them, there is no point in trying to improve the language itself because no one except philosophers are likely to ever use those improvements, and setting aside the problems eliminates the need for them.