It's amazing; with each caffeinated beverage I consume, I want to write another post!
I want to nuance my last post a little bit, since someone who didn't know any better might believe that I was simply bemoaning reductionism in disputes, and there is more to it. It's true that I think the main sin of spectrum-thinking and other like fallacies is one of oversimplification, and a failure to make proper distinctions. However, though making good distinctions is not the sole purpose of philosophy. We also have to subsume common elements into recognizable categories. Doubtless people can see me doing that when I try and get at the root of liberalism as a single phenomenon in the history of ideas. At least I make an attempt not to group disparate features capriciously.
The special problem of trying to form conceptual umbrellas under which to group disparate things is that these umbrellas will always be "extreme". For example, if I include under the heading of "liberal" the qualities of prioritizing the nominalist/modern scientific notion of knowledge, the flattening of values, and a mechanistic concept of the human body, my definition begins to differ little from the worst nihilistic atheist. It is a self-consciously stereotypical model; a banding together of characteristics which distinguish liberalism from its contraries, and thus allowing nothing of moderation in itself.
I don't believe that this is a bad practice, but it can get arguers into trouble. The most public recent example is the skirmish between Pope Benedict and radical Islam. Although the most violent reaction to his Regensburg speech was due to a misunderstanding, there was more moderate criticism of his speech from moderate and educated Muslims. At the core of this criticism is that Pope Benedict established a conceptual umbrella he called de-Hellenization, and under this category he assigned Protestant fundamentalism, and Islam--perhaps while failing to make proper distinctions as to the "Islam" he was speaking of.
I do not believe the Pope was wrong to make this argument. There is an iconoclastic strain in the history of ideas that pops up now and again, and seeks to rid religion of any pagan rationalizing. Moreover, that "dehellenizing" strain can be found prominently in the birth and spread of fundamentalist movements in several religions; Protestantism and Islam are two easily accessible examples.
What people probably missed was that the Pope was making use of a certain style of argument. I don't have a name for it. Maybe somebody does. But it goes like this: "See, there is this thing, X, and X is not very good, so we should avoid participating in X. Now see how these various things, A, B, and C, participate in X, and how this causes common problems we see in the world today."
The key word is participation. Whenever we construct a model or a category, under which we (rightly or wrongly) subsume various features, and then we make a comparison of that model to something concrete, we implicitly recognize that the model and the concrete thing aren't exactly the same. So when we say that A participates in X, we mean that A has features in common with X, and X is something we know about, so maybe X can tell us where A comes from, or is going to.
Using models and general concepts, and arguing based on them, is not a fallacy, but it is the stuff out of which so many fallacies are made: straw-men arguments, racism, stereotypes, and so on. I suggest that arguing from models and concepts falls into these fallacies when one of two things happen:
1. The concept itself is inadequate. The pieces of the concept have no essentially unifying quality, taking their apparent coherence instead from prejudice and groupthink. This is certainly the case with racism, but it is also the case with any provincial prejudices and irrational fears that nobody in this sin-torn world is ultimately free from.
2. The concept is adequate, but in the rhetoric of comparison, one fails to clarify that it is a model in which the concrete thing participates. Instead the model and the concrete fact are simply conflated, which leads to the accusation of stereotyping. An example. Suppose for the sake of argument that we take The Simpsons as an adequate model of a dysfunctional family. It would be one thing for a family counselor to draw comparisons between the actions of, say, the real father and the actions of Homer Simpson. "Do you remember that episode where Homer ...? Maybe you could learn from his mistake." It would be quite another for the counselor to say, "You, sir, are like Homer Simpson."
The thing about arguing based on general concepts, and using the language of "this participates in that", is that it incorporates both "black and white" thought and "spectrum thought". There are blacks and whites; there are models and concepts and discoveries that have some kind of reality, some stable inner coherence, such that they might even be simultaneously discovered by disparate peoples. But at the same time there is the fact that we never find any of these concepts in pure form within the concrete realm. So instead we say that A participates in X--and in saying so, we admit that this participation happens in varying degrees.
If anyone has observed that I have suddenly become very Platonic in my thought, then have a cookie.
I bring this out not to contradict my last post but to sort of bring it full circle. Yes, there is legitimate "binary thought" and "spectrum thought" but they never exist in isolation from the other and they do not excuse the preliminary task (that I emphasized in the previous post) of going through and making all the proper distinctions.