Logical analysis of ad hominem

Dean has cited the comments on the further debunking of his HIV/AIDS nonsense as, "profitable." As I read his response there, his position makes less and less sense to me. First, he accuses McBride of making the ad hominem fallacy for attacking Al-Bayati's credentials. Then he says:

I tried reading Orac's response, by the way, but it was so full of false accusations that I couldn't get through it. The man argues first and foremost through attacks on others' credibility. I do not, for example, "wave credentials" at anybody, not for my friends or detractors. I note only that credentials do matter, and that in any area of science, a qualified scientist should be respected as having the right to a dissent, even if it’s a minority position, AND, that one shouldn’t treat qualified scientists as kooks or people who can be dismissed with an airy wave of the hand. That is all I have ever said on the subject of credentials; Orac’s dishonesty in this leads me to the (admittedly ad hominem) conclusion that he can’t be trusted to be honest on anything else.

Is Dean's position, perhaps, that credentials are important, just not important enough to be critically examined?

What really caught my eye, though, was how he describes ad hominem arguments:

It's a logical fallacy that works like this:

1 Person A makes claim X
2 Person B makes an attack on person A
3 Therefore, A's claim is false

When I saw Dean using strict logical form, I got excited because it meant a chance to do something useful with careful philosopical reasoning.

First, we need to understand what a deductively valid argument looks like:

1 My brother is a cat
2 Cats have fins
3 Therefore, my brother has fins

Silly, but completely valid. Why? Though all three statements are false, if the first two were true, the third one would have to be true as well. Now, a fallacious argument:

1 George Bush is American.
2 All Texans are Americans
3 Therefore, George Bush is a Texan

The problem here is that Bush could be from California even if the first two statements were true.

This should make clear exactly what is wrong with ad hominem arguments, as defined by Dean. A person can make a claim and be attacked without the claim being false.

The following however, is a valid argument:

1 Arguments made by person A are likely to be flawed
2 Person A has made argument X
3 Therefore, argument X is likely to be flawed

This can be done with a few variations. The argument may be narrowed to apply, say only when A is talking about theoretical physics or when quoting famous scientists. The first premise may be established several different ways, such as with reference to bias or past mistakes. Different initial premises may also establish different probabilities of the person being mistaken.

The usefulness of such an argument can vary greatly with the situation. Suppose someone, after refuting Thomas Aquinas' first four ways, proceeded to attack the fifth in the following:

1 Thomas Aquinas' arguments tend to be flawed (see refutations of 1-4)
2 Thomas Aquinas made the argument known as the fifth way
3 Therefore, chances are the fifth way is flawed

Such an argument is valid, and may be all you need to put the fifth way out of your mind when debating someone more enthusiastic about William Lane Craig's Kalam argument. However, if there is a serious question as to whether the fifth way is sound, there is a much more certain way to decide this: direct analysis of the argument. This should be possible, assuming whoever you're debating has presented the argument itself, and is not merely saying, "I forget the argument, but at the time I read it it sounded pretty convincing."

Now consider this argument:

1 Chris Hallquist cannot tell the difference between a spoof and something said in earnest (see here, here, and the fact that I though OBJECTIVE: Ministries was real
2 Chris Hallquist has cited X as an example of something outrageous a fundamentalist has said
3 Therefore, X may very well be a spoof

Again, you may be able to get a clearer analysis by looking at the thing itself. However, if I didn't provide the source, or you look at the source and can't tell, such an argument is worth keeping in mind.

Now, bringing it back around to the original debate, legitimate arguments can be made involving Al-Bayati's credentials. For example:

1 Al-Bayati is not qualified to perform an autopsy
2 The cause-of-death findings of someone not qualified to perform an autopsy should not be trusted over those of a medical examiner
3 Therefore, Al-Bayati's findings should not be trusted over those of a medical examiner

Let me add that to say, "His credentials have been attacked, therefore the argument is ad hominem, therefore the argument involves the fallacy described by Dean" is every bit as fallacious as the original fallacy.

This also means that Dean is wrong to say his conclusion about Orac is ad hominem in the sense he laid out. He could form his argument this way:

1 Orac was dishonest in his response
2 Someone who is dishonest cannot be trusted
3 Therefore, Orac cannot be trusted

Valid, though as I stressed in my explanation of validity, a valid argument may have a false premise. Dean supports premise 1 by citing Orac's statement about waving credentials. To my knowledge, Dean only blogs and does not engage in face-to-face debates, so he has probably never picked up someone's diploma and waved it in the air. I think all concerned are aware of this, making it clear Orac used "waving" in a metaphorical sense. As to whether Dean has performed metaphorical waving, one might consider the time he responded to a critic solely by saying two Nobel Laureates argeed with him. Whether this counts as metaphorical waving is a difficult question, which I leave to my readers to discuss in the comments. However, I think there is room for Dean to considerably strengthen that first premise.