What is a wife?  From a philosophical point of view, it looks like the word `wife’ is a predicate and so should stand for a condition, presumably one that humans meet or don't meet  at times.  And so the first question is, which condition?  And then the next questions would be about the importance of the property, its relation to issues of equality, social structure and the like.

But the first question isn’t so simple.  When we read Marilyn Yalom’s book, A History of the Wife, it seems pretty clear that the word `wife’ (and the words in other languages that it translates) has designated different conditions  in different ages and cultures.  In many ages and cultures, being a wife was being a certain sort of property.  That wouldn’t go down well in contemporary America. A contemporary American  wife is legally her husband’s equal, and in fact often his superior educationally, financially, and in a lot of other ways.  So has the meaning of `wife’ changed?  Is it then just ambiguous?  Systematically ambiguous?  Is it not ambiguous, but nevertheless manage to stand for different properties at different times?  How would that work?  Is it context dependent?  Is there an unarticulated constituent hidden in their somewhere: a wife is a wife in a culture at time? 

These issues, which don’t connect too much with what we talked about with Marilyn Yalom, do connect with things I have been thinking about in connection with the problem of freedom and determinism. 

It seems to me that when philosophers think about words that have a big role in ordinary life, they need to distinguish three questions, which I’ll get to in a minute.  By words that play a big role in ordinary life I mean, in the freedom and determinism case, the phrases we construct with the word `can’, as in ``I can finish this blog in less than an hour,” which may or may not be true.  It seems to me very likely that these can-phrases, or their translation, were in use in human activity long before philosophers began worrying about what they meant, and that the  conditions that they stand for almost certainly played a role in human thought and activity before there were words for them.  And I suspect the same is true for the word `wife’.

In contrast, words like `free-will’ and `democracy’ and `social contract’, although they now enter into all sorts of conversations by non-philosophers, probably had their beginnings with reflective thinkers of some sort, theologians, philosophers, and their ilk.

Now, with the first kind of word, it seems to me that two ideas from twentieth century philosophy (you remember it), are very useful --- two ideas not always thought to be very connected.  One is Wittgenstein’s idea of a language game, and the other is the idea of direct reference.

Wittgenstein thought it was useful to think about the meaning of words by constructing primitive ``language games’’ in which the words are part of a small fragment of language that is integrated with a human activity, so one can see vividly the work the words do in the sort of primitive situation in which they came to have meaning, or could have.  The most famous is his builder-assistant language game, in which there are just four words, block, pillar, slab and something else, I don’t quite remember.  (Those who would like to delve into gory detail about this language game can go to my paper ``Davidson’s Sentences and Wittgenstein’s Builders, available at http://www-csli.stanford.edu/~jperry/phil.html.)

In this case, one can see that the words are used to stand for a condition or property --- I try to use `condition' for something perhaps rather complex, not something nature necessarily offers up as important independently of human activity --- that plays a certain role in the life of humans.  Humans know how to tell the difference between pillars and slabs, they use them for different things in building, and so forth and so on.  The words get meaning by being connected with this role; a pillar is the sort of thing the builder uses for certain purposes in certain building situations, and the assistant gets praise for bringing if the builder says `pillar’ but not if he says `slab’. 

This role might be thought of as providing a description that the builder and assistant might associate with pillars, so that pillar could be seen as the condition that all things have in common that meet the description.  But this is a pretty odd way of looking at it.  In the first place, the players of this language game don’t have any other words to use in their descriptions.  It’s their activity, not their conceptual apparatus, that is the key to the link between word and condition.   If they grow to have opinions on the subject of the difference between slabs and pillars, those opinions might not be accurate. The builder and assistant might have different, even inconsistent, opinions about the right way to define `slab' and `pillar'.  It doesn't matter, as long as they agree on the extension --- as long as the assistant brings the right sort of building block.  It is the role the condition plays in their activities that is important, not the opinions they may come to have about what is definitive of slabs and pillars that is important.  That’s the connection with direct reference.

So that’s the first question: what role that things play in human activities, especially the sorts of activities that might have been in place before the word, or any words, were in place, is the word connected to.

Now in the case of `wife’, it seems reasonable to suppose that people were linking up in male-female relatively long-term relationships long  before language in general or the word `wife’ in particular came along.  The words `wife’ and `husband’ get at two roles in these relationships, being the female member of such a relationship, and being the male member.

Then there is the question, what, if anything, do people think about the occupants of these roles?  What assumptions to ordinary folk make?  What opinions have theologians and philosophers and judges and others in various kinds of conceptual and legal authority come up with?  What further practices, that depend on some or all of these opinions, have developed in different times and places, and why?  It is the answer to this question that, in part, a book like The History of the Wife, tells us about. 

Given this framework, we can get a handle on the original problem, how can `wife’ stand for so many different conditions or relationships, and still in some sense have the same meaning?  There can be a basic role in a widespread and long enduring human activity that the word is associated with in many cultures and times, but the opinions about what goes along with playing the role can different from culture to culture and age to age.

But it is more complicated, because of the third factor.  Some of the opinions are codified in various ways that have some sort of official status.  There are laws in various places and time about wives, whether or not they are property, whether or not they can own property, whether there can be more than one to a husband, whether there can be more than one husband for a wife.  Laws, secular and religious, may establish an official meaning for a word that diverges from its root-role.  A man and a woman may marry, but never live together; they may live with other people, but still be husband and wife.  A man and a woman may live together and rear children, but not marry, so not be husband and wife.

This generates a new sort of question:  what should a wife be, in order that these various practices that have been codified in more and less informal ways, make sense, given one or another set of ideas about people and justice and the like?  Or, are the people that fall in the extension of `wife' as used, deserving of the treatment so codified?  If not what should be changed?  The extension (males can be wives, since there is no justification in sex and gender for not allowing them the privileges that come with being the spouse of breadwinning male?).  The practices (wives should not be treated as property, because wives are persons, and persons can't be property)?  To understand the dynamics of the issue, who and what is a wife, we would probably have to develop some account of how all of these factors fall into a sort of equilibrium, relative to the mores of a given society and a given age.

 Is that what's involved in philosophical analysis?

What a mess.