The dialogue I am interested in is between two persons whose religious views relevantly differ. The puzzle is that (a) such dialogue seems to actually happen--it seems that people are having a dialogue, both talking on and the same topic, exchanging views that make contact, even though (b) it seems that the two people mean different things by the words they use.
Let me give a couple of examples, in all of which one interlocutor is a Catholic and the other a naturalist, though many other combinations will generate similar examples.
- Marriage: The naturalist (typically) understands marriage as a social status, conferred by society, while the Catholic understands it as having an objective component that is at most causally and contingently dependent on the conferral of a social status.
- Papacy: The Catholic's concept of the Pope is such that it is a priori that the Pope is the successor of Peter (if Peter had no successor, then there is no Pope), while the naturalist's concept of the Pope makes it at best an a posteriori truth that the Pope is the successor of Peter.
- Baptism: The conferral of grace is a part of the concept of baptism for the Catholic--it wouldn't be baptism without it, but "invalid baptism" (which is related to baptism in the way in which a merely apparent oasis is related to an oasis), without it. On the other hand, the naturalist understands baptism as a human ceremony.
Now, one might think that productive dialogue requires that everybody use words in the same sense. Indeed, the naturalist could accept the Catholic's concepts of marriage, the papacy and baptism. She would, however, have to then say that no one has ever been married or baptized, and she might be unsure whether there is a Pope. In practice, however, the naturalist doesn't typically talk this way. She talks of marriages and baptisms as if their occurrence were a fact, and she is as certain as any Catholic--and more certain than some--that Benedict XVI is the Pope. Such an adoption of the Catholic's concepts of marriage, the papacy and baptism would require circumlocutions on the naturalist's part that are not only awkward but involve a breach of courtesy: "A friend invited me to what he thinks is his daughter's baptism"; "...the so-called Pope Benedict..."; "Tomorrow, Eve and Patrick will seem to get married." This is not what typically happens in productive dialogue. So, interestingly, it would be inimical to productive dialogue that the naturalist accept the Catholic's concepts.
Perhaps a more productive approach would be if the Catholic adopted the naturalist's concepts of marriage, the papacy and baptism. In many circumstances, that would work out just fine, and in fact in ordinary speech, Catholics will do this with the word "marriage". Thus, very few Catholics--in part due to issues of courtesy and prudence--refer to "Henry's putative wife" in a case where Henry was married to another and divorced, and the previous spouse is still alive.
But such adoption of the naturalist's concepts is not a hard and fast rule. Indeed, if the dialogue makes serious use of these concepts, it may be quite necessary for the Catholic to make, say, a distinction between a socially accepted anti-Pope and a Pope, or between a merely putative marriage and a marriage, or to explain why Fred upon converting from Mormonism was baptized, while Jenny was not baptized after converting from Calvinism.
So it seems that sometimes we have a dialogue, and apparently a productive one, between two people who fairly consistently use words in different senses. That this is possible is puzzling. Yet it is actual, and hence possible.
How to solve the puzzle? I am not sure there is a general solution. I think in a number of cases, the two sides use concepts that have sufficient overlap in extension that helps make the dialogue productive. Thus, while in my sense of "married", the naturalist does not believe that Fred is married, nonetheless she says (using her sense of the word) that Fred is married, and I say (using my sense of the word) that Fred is married. (Curious linguistic fact: in ordinary speech, after I said that Fred is married, the naturalist would say that what I said was true, even though the proposition I expressed is one that she thinks is false.) We can then go on to debate whether Fred's being married consists in something more than a social status. It would seem that we are talking at cross-purposes: on my view, by definition marriage consists in something more than a social status, while on the naturalist's view, by definition it does not. However, the partial agreement on extension means that there is a meaningful question here: namely, whether Fred's acquiring the social status of "being married" was accompanied by any further non-natural effect.
Alex,
This is a really interesting post. I think I agree with where you come down, namely that the overlap in extension, in most cases, is doing a lot of the work. This allows (somewhat surprisingly) a lot of *communication* to go on even while different concepts are being used. If we say it allows *dialogue* then this seems to imply that we can have productive dialogues *about marriage/baptism/the Papacy* while holding different concepts of these things, and on this point I am skeptical. One of the first things we need to do in such situations, I think, is define terms.
However, the overlap in extension is no accident. We might put it like this. There is in each case a set of easily identified instances (of marriages, of Popes, of baptisms) which are the starting points for both parties. Then each party develops a theory of what unites the instances. These theories differ slightly in extension but are constrained by the numerous paradigm cases that form the starting point for the concept.
If disagreement on the paradigm cases becomes rife, then the sort of convenient equivocation you have identified becomes impossible. So, for example, I am not sure that "sin" can be used in serious communication between a Christian and a non-believer at this point.
Perhaps our use of terms like "marriage" or "sin" is fluid in a way. If we say, for instance, "Glenda and Gary's marriage", perhaps in some cases we, as it were, point linguistically. We direct attention to the two and the relationship between them (whatever precisely that is), and with that entity picked out, we can then make our point about it. On uses of this sort, we don't mean to say that their relationship satisfies any thick and controversial concept of marriage. Instead we simply mean to fix a referent about which we can then dispute.
I of course don't mean to say that in no context do we use a thick concept of marriage. When our interlocutors agree with us about that thick concept, perhaps we do. But I suspect that we sometimes fall back on this thin, quasi-indexical, sense.
Hi Alex,
Could you say something about how the sort of issues you are talking about differ from familiar issues concerning names and natural kind terms? Suppose that you think water is composed of H2O molecules. Suppose that I think water is composed of XYZ. We can still have a conversation about water. How is what you are talking about different?
One might think that what allows for dialog between naturalists and Catholics is that they are referring to the same thing when they use terms like 'baptism'.
The Causal Theory of Reference delivers this judgment. Here is how: Suppose that John performs an act of baptism. Someone points to it and says "That is a baptism." Then that person speaks to other people and uses the term 'baptism'. Then those people can use 'baptism'. They associate different descriptions with 'baptism'. Some include supernatural stuff. Some don't. But in virtue of the causal chain running from John's performing the baptism to each speakers use of the term 'baptism' they refer to the same thing.
Perhaps I'm missing something.
HW:
The sin question is interesting. One of these days I'd like to try to understand the concept of sin better. It's something thicker and richer than just the concept of a moral misdeed. Maybe it's the concept of a moral misdeed that offends God, but I think that may not be rich enough.
DJ:
The causal theory of reference move is a nice one. It's related to, but more sophisticated than, my suggestion about significant overlap of extension.
One problem is that to get a causal theory of reference to typical second order entities (relations, properties, etc.), you will likely need more than one instance as the cause of the concept. If I define "marriage" by pointing to Sue and Frank and saying that marriage is the symmetric relation between them, my definition is surely indeterminate, because typically there will be many symmetric relations between them, such as, e.g., friendship, conspecificity, similarity of age, cohabitation, being of opposite sex, tennis partnership, etc. You'll need to point to a bunch of married couples.
But there might not be enough overlap on paradigm cases, especially in the marriage case these days. Given the percentage of people who are divorced and putatively remarried, a significant proportion (though not a majority unless the sample is unrepresentative) of the naturalist's paradigm cases of marriage will actually be couples that according to the Catholic are not married. (I am not assuming that every divorced and putatively remarried couple is not validly married on the Catholic view. E.g., some may now be validly married because their first putative marriage was invalid.)
The second problem is this. Think of, say, baptism as involving a natural and a supernatural component. The natural component involves sprinkling of water, saying words, etc. The supernatural component involves grace. Now, if the naturalist view is right, then there is not too much of a problem with gaining reference to baptism in the causal way. But on the Catholic view, the supernatural component of x's baptism need not have any effect on y's thinking about x's baptism (it might--it could be that the graces become manifest in x's actions to y--but it need not). If not, then only the natural component of baptism will stand in the causal relation to y. But then it seems that on the causal theory, y's concept is not of baptism, but of the natural component of baptism. And that's not in fact what is the case.
One might try to get out of this by invoking naturalness, saying that one gains reference to the natural unit that contains the cause. It's not clear, however, how much something sacramental counts as a natural unit. This is going to be quite tricky at least.