The Onion reports:
ABUJA, NIGERIA—At a celebratory press conference Monday, President Olusegun Obasanjo announced that Nigeria's troubled but oil-rich city of Warri has been chosen to host the 2008 Genocides.
I don't know whether anyone has brought this up before, but since I haven't come across it, I thought I might as well jot it down here...
People have often wondered about the biological function of the kind of deep, lasting grief that accompanies events such as the loss of a child. Why is it so debilitating, and why can it be so debilitating for such a long period of time? Wouldn't it make more biological sense to let the person get on with their life and the business of propagating their genes?
In Stephen Pinker's book How the Mind Works (which would more accurately, but with less impact be called, "An Overview of What We Currently Know About How the Mind Works") he notes that such grief seems to act as a kind of deterrent - basically, knowing how awful it feels acts as a major deterrant to any behavior that might lead to those kinds of circumstances arising.
It occurred to me that perhaps the deterrance is not just for you -- but for those around you, and that this factor could help explain why it is so highly debilitating (if in fact this actually requires any separate explanation). Not only do you experience how awful it is, but others can also see how awful an effect it can have on you. And since in the times when evolution was at work we lived mostly in tribal groups with large numbers of relatives, Kin selection would explain how this kind of grief could evolve to serve "both" these purposes.
Perhaps just the right impact on the grief is where it has a large precautionary effect on the other kin members at the expense of really debilitating the grieving person? (i.e. this is the level of grief that results in greatest chance of those genes propagating, because the least number of people carrying the gene make the mistake that led to the grief).
Some people (and I don't think they're the only ones) talk about representations as if something was only a representation if it was in some fairly explicit form that can be read by some process which then interprets its meaning.
For example, consider the rules of grammar that our brains respect (to a certain degree) in the production of speech. Under this view of representation, there are two options for how these rules exist in our brains: either these rules are represented in the brain, or they are in some sense built-in to the brain without being represented. In the latter case the system in some sense knows the rules, but it doesn't explicitly represent them -- the brain just works in a way that respects the rules, just as a calculator doesn't explicitly represent the rules of arithmetic, but just operates in a way that respects them.
This distinction between things that are represented and things that are 'built-in' sounds fine on the surface, but I'm suspicious of there being a real distinction there. Certainly the distinction makes intuitive sense. It's easy to imagine an explicit representation there that is to be read off and used. And its also easy to imagine the other case, of a system with no representation, that just works in a way that respects the rules. However, the thing is that a representation is represented by some encoding scheme. You can represent something an unlimited number of different ways (of course, different ways will have different properites, such as how long it takes to read the representation, how much space it takes up, etc etc) and in each and every case what is important is that you have the right means of interpreting the representation.
And amongst this unlimited variety there are encoding schemes that more explicitly represent the thing and those that do so less explicitly. So my question is, is there just a spectrum here, from very implicit representations right up to very explicit reprsentations? And is it possible to to make any hard divisions of this spectrum? I strongly suspect that it is not possible to make any hard divisions or distinctions.
In the case where the brain simply respects the grammar rules but does not explicitly represent them would still be a case of the brain representing those rules, but just implicitly. Taking another tact on this, I don't see how the brain can "know" the rules yet not in any sense have a representation of them. With the right means of interpretation we can still 'read' the implicit representation (not that we, as people trying to understand how the brain works, necessarily have the means to determine this means of interpretation). In other words, some aspect of the system's organisation implicitly encodes the rules, and the way that that organisation interacts with and influences the behaviour of the rest of the system consitutes the interpretation of those rules.
It may seem a bit of a stretch to call such implicit cases 'representation'. But if what I'm suggesting is correct, then there's no hard distinction between "explicit representation" and "implicit representation", and thus the common meaning of representation is making a hard distinction that doesn't exist. Thus, if we are looking for technical accuracy (and of course we are often not) then we either have to modify the common meaning of the word or find another term that can have a more accurate meaning.
