Saturday, 20 May 2017

On Whose Authority..?

"If it disagrees with experiment, it’s WRONG. In that simple statement is the key to science. It doesn’t make any difference how beautiful your guess is, it doesn’t matter how smart you are who made the guess, or what his name is… If it disagrees with experiment, it’s wrong. That’s all there is to it." - Richard Feynman

By the time we get a bit into this outing, that's going to seem like a pretty ironic thing to have done, to quote my favourite physicist saying something, but what he said, and the direct and pithy way he expressed it, goes to the very heart of what I want to discuss.

What Feynman was talking about there was one of the central principles in science; falsifiability. That's not the subject of this post, however. Today, I want to cover a topic that's been waiting to be vented for a while, and this morning I met the proverbial straw, so here it is.

There's a very common group of fallacies that we've touched on briefly before, and there will be a little repetition, but here I want to examine them in more detail. That group comes collectively under the banner of the 'genetic fallacy', and comprises all fallacies that are rooted in the source of the argument.

What finally drove me to rant about this was two occurrences on Twitter in which several completely different iterations of the genetic fallacy were committed by the same person, finally to be capped off by yet another iteration.

The key portion of what Feynman said that has relevance to this is "it doesn’t matter how smart you are, who made the guess, or what his name is…". What he's expressing there is a beautifully brief debunking of the genetic fallacy in all its forms.

So just what is the genetic fallacy? In essence, it's any instance in which an argument is accepted or rejected based only on who said it, generally based on some characteristic thereof, whether real or merely perceived.

The most common form of the genetic fallacy is the argumentum ad hominem. This fallacy is the lowest form of discourse, and is committed when, rather than actually address the content of the argument, an insult is hurled at the arguer. Often mistakenly understood to be any insult in an argument, the fallacy is only committed when the insult is the only response to an argument, and no attempt to actually refute the argument is made. To clarify this, consider the following two examples.

1. Your argument is invalid, you moron, and here's why... (goes on to explain flaws in argument).

2. You're a moron, so your argument is wrong.

The former, while containing an insult (setting aside for the moment whether a true statement can really be an insult), is not fallacious, thus does not commit the argumentum ad hominem. In the latter example, however, the argument is dismissed purely on the basis that the arguer is a moron (or, at least, perceived as such). This fallacy can best be criticised in the simple adage that even a broken clock is right twice a day.

Probably the second most common form of the genetic fallacy is the argumentum ad verecundiam, or appeal to reverence or authority. Those of us engaged in counter-apologetics to any great degree will most regularly encounter this in some form such as "Scientist X believes in God, therefore my position on god is well-supported". Just such an instance was one of those I encountered earlier today, in which the apologist was saying that Newton, among others, was a young-Earth creationist. To my certain knowledge, Newton never expressed any views on the age of the cosmos, and he certainly didn't adhere to the brand of creationism propounded by this tweerp. All other considerations aside, Newton thought the trinity was nonsense, and it required a special dispensation for Newton not to have to take an oath of conformity in order to take up the Lucasian chair at Cambridge, the same chair later held by Paul Dirac and Stephen Hawking, among others.

Newton is an instructive example for our purposes here. Among other things, Newton was an alchemist and, in fact, dedicated considerably more of his time and effort in attempting to find the 'philosopher's stone' than he did in physics. Certainly, the experimental methods he developed in his alchemical studies later found great use in both chemistry and in science generally, and they formed the backbone of the experimental methods we use today. That doesn't mean that alchemy wasn't complete bollocks, of course. This should be a warning fairly audible to us all that authority is fleeting.

I should say a word about that quotation at the head of this missive, because it might seem that I've contradicted myself in quoting Feynman here, especially in the context of a warning about the fallacious nature of reliance on authority. 

There's a distinction that should be drawn here, vaguely reflective of Euthyphro's dilemma, but much easier to resolve. Did Feynman say this because it's true, or is it true because Feynman said it? The answer is, of course, that he said it because it's true. The reason I quote him here is not that it was him who said it, but because it was a very pithy way of stating the issue, and to state it like that and not to credit him with saying it would be plagiarism.

There are those who suggest that there can be a non-fallacious appeal to authority. We looked at this in Argumentum ad Verecundiam and the Genetic Fallacy, in which we looked at consensus in science. Apologies for the repetition, but it's useful to look at the example I gave there.
So why is appealing to authority a problem and, more importantly, why does the charge of fallacy not stick in this instance? Let's look at an example that highlights the fallacy.
P1. Richard Dawkins says the universe came from nothing.P2. Richard Dawkins is a respected scientist.C. The universe came from nothing. 
There are those who would see no problem with that, though I doubt any of my readers will fail to spot the glaring flaw. Yes, Dawkins is a respected scientist, but his field is not cosmology, it's ethology. It is of no more moment that Dawkins said this (he didn't, by the way) than that Bill O'Really thinks we can't explain the tides. 
Let's try another one.
P1. Lawrence Krauss says the universe came from nothing.P2. Lawrence Krauss is a respected cosmologist.C. The universe came from nothing. 
Surely we're on firmer ground here? 
Frayed knot! Yes, it's true that Lawrence Krauss is a respected cosmologist and, yes, he is an expert in the relevant field. The problem is slightly more subtle here, but it's that Krauss is one cosmologist, so it's of no more moment that Krauss said this (he didn't, by the way) than that Isaac Newton thought that gravity propagated instantaneously.
In that post, we went on to look at the process of peer-review, in which the research is looked at in great detail by experts in the relevant fields and, if it survives this process, it enters the realm of mainstream science. In reality, this process never ends, and all conclusions are tentative, but that's not important for our purposes here. The takeaway is that no single physicist, nor even any group of physicists, or scientists in any field, can be taken as a valid authority for the purpose of drawing categorical conclusions.

It's critical that we're aware that only conclusions that can be supported by validated research hold any authority, and that even that authority is fleeting. Coming back to what Feynman was talking about, even our best-supported theories must be subject to potential falsification, or they can't be deemed scientific. This is the demarcation problem, discussed at length in earlier posts, and solved by Popper. If it isn't falsifiable in principle, it isn't scientific. Always remember, anything that explains everything explains nothing.

Another common example of this particular iteration is something along the lines of 'Anthony Flew was an atheist, yet he came to believe in god' or, worse, 'even Darwin doubted his own theory', generally coupled with some quote-mine.

Here's the thing: It wouldn't matter to me if Darwin, Flew, Russell, Spit the Dog or even Richard Dawkins knocked on my door tomorrow and told me that they'd been wrong all along and that evolution was a false idea. What would make a difference would be that they could present evidence that it was false. 

There's another very common example of the genetic fallacy that warrants a look, namely argumentum ad populum, or appeal to popularity. This one's fairly straightforward to debunk, and it's been done time and again. The most common debunk of this is itself rooted in error, namely the idea that once everybody thought the world was flat. I addressed this error in a fairly comprehensive manner in DJ! Spin That Shit! in which I showed that we can't, in fact, show that anything other than a minority thought this. However, the point of the debunk is still valid, namely that it doesn't matter how many people think something to be true, it only matters that they can show it to be true with rigorous research diligently documented and robustly reviewed by people who genuinely understand the research.

There is, it should be said, a popular principle known from the Latin as vox populi, vox dei, which translates directly as 'the voice of the people is the voice of god', and which expresses the idea that the answer given by the majority is most likely to be true. There is some weight to this idea, but it has to be taken to apply only in a very narrow set of circumstances.

Probably the most familiar example of this appears in the popular quiz show Who Wants to be a Millionaire, in the lifeline known as 'ask the audience'.

For anybody unfamiliar with this show, the idea is that a contestant must answer fifteen questions to win a million (insert local currency here). Those who haven't seen it directly may have encountered the show in Danny Boyle's Slumdog Millionaire. On the way to answering these questions, a few lifelines are available to negotiate the tricky bits, in which the contestant can take advantage of three get-out clauses to avoid a question they don't know the answer to. They are phone a friend, 50/50 and ask the audience. 

Now, I have a particular weakness when it comes to quiz knowledge, namely popular culture. I know that, if a question were to come up about, say, boy bands, or soap operas, I'd probably flounder. These are perfect questions for 'ask the audience', not least because popular culture and this sort of ilk are exactly the kinds of question that you'd expect the majority to know the answer to. Ask them a question on quantum field theory, however, and they can't be trusted.

There's another way that vox populi is likely to fail, and it's rooted in a principle elucidated by David Dunning and Justin Kruger known popularly as the Dunning-Kruger effect. We discussed this briefly in Deduction, Induction, Abduction and Fallacy. This effect deals with competence versus perceived competence in a given area. 

I had a discussion some time ago about 'the smartest person in the room' when somebody presented a deepity by John C. Maxwell that ran thus:
The smartest person in the room is never as smart as all of the people in the room.
One of my favourite tweeps, who we've met in earlier posts and who we'll be meeting again in a near future post directly related to her field, asked the pertinent question:
Doctor Weinstein's tweet is, of course, what brought it to my attention in the first place (she always finds me the best WOTI*), and it immediately raised alarms.

What the research by Dunning and Kruger showed is not only that incompetent people tend to overestimate their competence, but competent people tend to underestimate theirs. In a situation in which a question is being asked, then, it may be the case that the smartest person in the room knows the correct answer, but is swayed away from it by Dunning-Kruger kicking in and, as a result, that person being won over by the popular opinion, which is wrong. This means that, even if you're counting the smartest person in the room as part of the 'all', there's good reason to suppose that the smartness of the all is significantly less than that of the smartest person in the room.

This is exactly the sort of platitude we should beware of, and it brings me to another iteration of the genetic fallacy; the appeal to intuition. I'm not going to go over this ground again here, not least because I'm conscious that this entry is getting long again. Use the search box on the right with the keyword 'intuitively' and you'll find at least one example of intuition being debunked. There are things that intuition will tell you are not true whose untruth would result in our not existing.

I'm going to deal with one more example, and it's really two in one. This is a reverse of the argumentum ad verecundiam,  and it commits another genetic fallacy along the way, namely 'poisoning the well'.

In this particular instance, an apologist asserted that there was 'volumes of evidence' and 'observable and measurable' that pointed to a deity. I, of course, suffering one of the planet's worst cases of SIWOTI syndrome, couldn't let that pass, so I challenged the apologist to present a single bit of observable, measurable evidence that pointed unequivocally to a deity. What did I get? This:
I, of course, pointed out that the first was an argumentum ad populum, but queried the second. I'm aware of the fine structure constant, a constant dealing with the strength of electromagnetic interactions between charged particles, and 'fine tuning', which I dealt with at length in Careful With That Dial, but I'm not aware of any 'fine constant', and suspected a misstatement or false conflation. When I queried this, I was presented with a Wikipedia link on fine tuning, which was telling in itself. Bear in mind that this person states in his bio that he's a retired physicist, a claim I suddenly found less than compelling.

At this point, the apologist decided it was time to question my credential. Let me be clear, although I've never been anything other than entirely clear about this; I have none. That's right, I make no pretence at having had any formal education in physics, or in any other field, in fact. I'm entirely self-educated, and my knowledge and understanding comes only from reading and from discourse with relevant experts. As it happens, I don't think I understand nearly enough, but I understand enough to know what the currently available data supports, and it in no wise supports a deity. 

What the apologist was attempting to do here was, of course, poison the well. This fallacy is a specific iteration of the genetic fallacy in which a characteristic is constructed specifically for the purpose of dismissing an argument.

Here's the thing: Not all education is formal, or comes with pieces of paper attached to them benchmarking educational achievements. Much learning is done simply, as Feynman put it, for the pleasure of finding things out. The history of epistemological progress is littered with the unlettered. Certainly, formal qualifications tend to show an aptitude for learning, but the absence of them doesn't necessarily indicate a lack of said aptitude.

There's a man famous among mathematicians, Srinivasa Ramanujan, just to name one example that came to mind. This man was born in India, and had no formal education. He discovered an old book of mathematics, and absorbed it all. He went on, with no formal education, to make significant contributions to number theory, mathematical analysis and infinite sets, among other things.

Edited to add:
There's an instance of poisoning the well that is fairly ubiquitous, and I had meant to include it, but I overlooked it in my eagerness to complete, so I'll insert it here in an indentation.
Often, one comes across an accusation that 'you only reject god because you want to sin', or some other such nonsense. This is another iteration of poisoning the well, known in logical circles as 'appeal to motive'. As can be seen readily, there are several possible reasons for rejecting claims of existence of deities, such as simply not finding said claims convincing. This is not only guilty of the genetic fallacy, it also commits another fallacy, namely 'affirming the consequent', a formal fallacy of the form:

P => Q, Q
∴P

 Or, P implies Q, Q, therefore P.
To see the form of this, we can look at the following argument:

P1. All men are mortal.
P2. Hitler was mortal. 
C. Therefore, Hitler was a man.
It's fairly easy to see where this goes wrong, not least because men aren't the only things that are mortal, and this conclusion disregards all of them.

I get it, I really do. It's easy to follow your intuition, or to accept the word of an authority. It's important to consider the source in any situation, but it's also important to consider that even the best source can be wrong. When we put our faith entirely in the hands of an expert, and don't consider that her expertise may not apply, or that the underlying assumptions upon which the expert's opinion is based might be wrong, we're miring ourselves in the genetic fallacy.

Seriously, don't take my word for it...

*Wrong on the internet, as in 'someone is wrong on the internet' - SIWOTI - a possibly fatal and definitely incurable syndrome that doesn't let me leave bollocks unchallenged.