In my previous paper “The Search for Universal Definitions” I argued that a definition that would be agreed upon by everyone would be impossible to find, and thus the existence of an objective meaning for words, one that can be considered constant among all people, or even most, should be considered unlikely. One reaction to this kind of remark is to assume that this implies argument and discussion would be impossible due to an inability to communicate unambiguously. After all, how can we agree to make a just city if we do not agree on what justice is? More radically, some would assume that this makes the truth of statements relative and without absolute validity if two people can disagree about whether an action is just and both be right. However, observation alone informs us that people can and do debate issues such as justice, and sometimes people even change their opinions in light of a strong argument, implying that communication can take place and that it is possible for some views to be “better” (closer to the truth or better supported) than others. This is possible because the word and the concepts underlying it, that is the meaning, can be distinguished from each other. If we are willing to discuss the concepts instead of the words we label them with, it is possible to communicate perfectly clearly, as well as retain the notion objectivity regarding the truth of statements.
Some of these matters concerning relativism arise from a poor understanding of what a definition is. The problem is that often the word and the concepts that surround it become confused. If we have a definition that states that something is X if and only if it is A, B, C… we have not identified the conjunction of these properties as being X by nature, but we have just labeled as X, the category formed by the conjunction of A, B, C… The category itself is the same no matter what we label it. For example, let us consider the principle that one should keep their promises. Some people would label this as an example of virtuous action, and some as an example rational self-interest, but whatever we are talking about it is clearly the same action. In this case, the category formed by “actions that result in the keeping of promises” does not change if we call it one thing or another. The case of more abstract entities is similar. For example, several groups may have a different concept in which they all call justice. In this situation it is not that some or all of the groups misunderstand justice, or are somehow incorrect; rather, they are talking about different things, different categories, and simply happening to use the same term to label them. There is nothing special about the word itself that dictates exactly what concept or category it should label, thus one group’s use is as valid as another’s.
At first, a separation of a word, the label, and the concepts it invokes in our mind, the category, may seem impossible, because when we think of the word, we reflexively think of the things that we apply it to. However in a way we had already achieved such a separation when attempting to create a definition. An ideal definition presents the meaning of the word without using the word itself. Thus we could, in theory, discuss justice without ever invoking the word “justice” itself if we use the description, which we were previously calling the definition, in place of the word. In such a discussion we no longer have to worry about what exactly is understood by “justice” since we have replaced it with an unambiguous description of a category of things. We can call this category a conception of justice, which is what one person thinks justice means, and if arguments are formed around this description instead of the word “justice”, they will not be misunderstood.
Let us illustrate this principle by examining how two philosophers with different conceptions of justice might attempt, that is they both label two different types of thing with the word “justice”, to arrive at a common conception of justice which they can base both their further theories off of, and thus avoid disagreements or misunderstandings concerning the meaning of “justice”. First, they would have to lay down some guidelines as to what justice is, in a very generic sense, without committing to one definition or the other. Let us say that they agree that justice moderates conflicts between people, and that justice is in the best interests of society, so that a just society will prosper. This describes their shared conception of what justice should be, even though they might disagree on finer details. Already we can see that some conceptions of justice are excluded, such as rule by the strong, which would create a weaker society overall. This description assumes that there is some agreement, but what if there isn’t agreement even on these basic ideas? What if one philosopher believes justice really is rule by the strong? In that case their discussion would not be called a discussion of justice, but rather some unnamed principle found in prosperous societies. Later, one might argue that this principle is justice, but they could at least agree on conclusions drawn about this unnamed principle.
From these shared principles observations, justice can be drawn, such as that theft is unjust and that it is possible for a society to encourage just behavior. It is possible that additional statements such as “justice is a virtue” may be introduced. These statements will first need to be reduced to their own general versions, such as “justice is beneficial to both parties”. Even so, such a statement should not be admitted unless both agree to it, or if it can be shown to follow from earlier statements. Through this process what I call a shared conception of justice emerges.
It is through the use of this shared conception of justice that both parties can communicate concerning justice. Let us label this shared conception of justice “justiceS”, and the private definitions of justice held by our two philosophers to be “justiceA” and “justiceB”. Instead of discussing their own private notions of justice, as they had been up to this point they now discuss only justiceS. Thus any conclusions drawn from this discussion will be shared by both parties. Likewise in such a discussion both parties will actually be talking about the same thing justiceS, instead of two similar, but different concepts, justiceA and justiceB. Of course, in their day to day discourse they will likely retain their old usages of justice, and may not even use the new term “justiceS” at all, the point of their creation of this new terms was not to change their use of “justice”, but to create a new term which would mean the same thing to both and which could perform many of the same functions. In this way, unambiguous communication can be established between people who use the same word in different ways.
Despite its appeal in terms of the possibility for objective communication, some may feel that such a treatment of the meaning of words opens the door for conventionalism or moral relativism. However this description of concepts as being divorced from the words we label them with is in fact a way accept the idea that words are in some way conventional without allowing that the truth itself is a matter of convention. The justification for this is that at some level the truth, and true statements, can be expressed in primitive terms, which are understood in the same way by all people, or at least there is agreement concerning their use. The truth of a statement can then be evaluated in terms of these fundamental ingredients, and thus will be acknowledged as true even if they label the statement in different words. Thus what one person calls a theory of virtue another may call a theory of rational self-interest, but at some level they are both the same, and both equally valid, if it is a well supported conception. For example, let us take two different societies, with moral codes and customs that are complete opposites of each other. Each one of these societies would label the other as evil, or at least misguided, and would call themselves just and virtuous. If we accept that there are no universal definitions we cannot point to one of these two societies and assert that it is the virtuous one while the other lacks virtue. Instead we must accept that both societies are using justice and virtue in their own internally consistent way, and that it does not mean that one of them is misunderstanding the term. However we are not forced to accept a position of moral relativism regarding these two cultures. Instead we can judge them on grounds of individual happiness, prosperity, or any other criterion whose definition is not in dispute, and on these grounds we can judge one to be better than the other.
Thus by giving up our reliance on an objective meaning for words I do not think that our discussions become relativized or somehow less based on objective fact. If you perceive certain actions as just, and believe that this perception is somehow based on objective fact, then it is easy to avoid relativism, but it is also extremely hard to change your mind concerning something which you simply perceive to be just or unjust. On the other hand, if you separate labels from what they are describing, it is much easier to accept that your instincts as to what is right and wrong can be mistaken, without being forced to accept that there are no standards. In fact, getting rid of the idea that words have objective definitions makes apparent the division between the word and the meaning, allowing us to more easily have a discussion concerning the meaning rather than the word, which is after all, what really matters.