Posts tagged Logic

Re: Thought for the day, 1: AI on Turing machines.

I think it’s possible that algorithms and computability are simply the wrong framework for understanding human intelligence or creating artificial intelligence.

They probably are the wrong framework. But I don’t see how that helps you with the halting problem. The issue with the halting problem isn’t that a solution is really hard, or that it’s ill-suited to the methods of computer programming. It’s that the notion of a correct general solution to the problem is, provably, internally contradictory. There’s nothing that could count as a correct general solution to the problem.

Here’s why. The halting problem is provably unsolvable because if it were solvable by any means, its output could be used to construct at least one algorithm which would be logically incompatible with the correctness of the solution to the halting problem. For a brief discussion why, see here. Basically, any proposed solution to the halting problem could be used to construct a “fink” function, which will do the reverse of what the halting problem says a given function will do when given itself as an input (so that, when you call FINK(FUNC), execution will halt if FUNC(FUNC) would loop forever, and will loop forever if FUNC(FUNC) would halt.) But then, when FINK(FUNC) gets itself as an input — FINK(FINK) — there is literally no possible answer to the question of what it will do. If it halts, then it doesn’t halt; if it doesn’t halt, then it halts. Hence, whatever your solution to the halting problem said it will do, it won’t do that. Hence, whatever your solution to the halting problem is, it’s wrong in at least one edge case.

Note that it does not matter whether the solution to the halting problem is accomplished within a Turing machine, or whether it is accomplished by some other means. Suppose the method for solving it is to print out a copy on a human being’s printer, and then wait for the human being to input TRUE or FALSE, and send the human being’s response back to the function to use as the return value for the halting function. Even so, you can still construct the fink function using that external input, which means that there just is no right answer that the human being could possibly return: whatever answer she returns, the program will do the reverse of what she said it would do.

The problem is not that finding a method to solve the halting problem is really hard, or limited by available resources or conventional computer architectures; it’s that the mere existence of a solution would necessarily entail an edge case where the solution cannot possibly be correct. A human being can’t do it any better than a computer can.

Re: Why We Fight (the Power)

Peter:

The “strategic-thickness,” “consequence-thickness,” “application-thickness,” and “grounds-thickness” arguments strike me as pretty insubstantial, to the extent I understand them. The grounds-thickness argument, for example — “Sure, private hierarchy is logically consistent with libertarianism, but it’s weird!” — seems like an assertion, not an argument.

Peter, are you referring here to the paragraph on authoritarianism in Libertarianism Through Thick and Thin, under the heading of “Thickness from grounds”, which begins “Consider the conceptual reasons that libertarians have to oppose authoritarianism, not only as enforced by governments but also as expressed in culture, business, the family, and civil society. …”?

If so, I’m not surprising you find the argument unsatisfying, because that’s an extremely elliptical capsule version of the argument. It’s intended to illustrate the kind of argument that you would make for a commitment from grounds, not to give a full-on account of the argument for libertarian concern with non-coercive authoritarianism. A fuller version, with the details tricked out, would require a lot more space than I had available in that part of that particular article (which was written for print in The Freeman, and hence subject to constraints of length, and which was primarily about the varieties of thickness, not primarily about making the case for all the details of my own particular thick conception of libertarianism).

There’s a bit longer discussion of the same topic in my “Liberty, Equality, Solidarity” essay in the Long/Machan Anarchism/Minarchism anthology (particularly if you include, as background, the section on equality), which you may or may not find more satisfying.

Whether or not you find it more satisfying, though, what I’m more interested in is whether or not you accept the form of argument discussed. Specifically, an argument in which the arguer demonstrates 1. that the best reason to be a libertarian is some foundational principle X (Aristotelian natural law, rational egoism, Jeffersonian political equality, whatever your view may be); 2. that principle X implies not only that libertarianism is true, but also some other consequent, Y; and, therefore, 3. a libertarian, qua libertarian, has reason to believe in Y as well as libertarianism, even though denying Y is not inconsistent with libertarianism per se, because denying Y would be inconsistent with the reasons that justify libertarianism. (Hence, as I say, libertarians can reject Y without being inconsistent but they can’t reject it without being unreasonable.)

So, do you accept that form of argument as a legitimate one? If so, then great; that was the main purpose of the discussion, and presumably also the main purpose of Roderick’s link to my essay. If not, then what’s the problem with it?

Re: A Spontaneous Order: Women and the Invisible Fist

  • Jerry: “If she is wrong and mischaracterizes the causes, what does that say about her conclusions regarding the effects?”

Nothing at all. If Brownmiller advances a false theory of the form “X causes R” (N.B.: I’m not conceding that her theory about the causes of rape IS false; nor am I insisting that it’s true; my position is that it’s not salient to this discussion whether it’s false or true), and then advances another theory of the form “R causes P,” based on an independent argument that doesn’t refer back to the first theory, the falsity of the first theory tells you nothing at all about whether the second theory is true or false, and nothing at all about whether the theory well-grounded or ill-grounded. What will tell you something about the merits of the second theory is a consideration of the independent arguments that are given for it.

  • “F => F is True. F => T is True.”

If you mean the arrow here to express a material implication, that’s an accurate description of the truth-values of material implications with false antecedents. But what has any of this got to do with the comments you’re trying to respond to?

There is no argument that I made, or which Susan Brownmiller made, in which her theory about the causes of rape is the antecedent in a conditional of which her theory about the effects of rape is a consequent. The theory in which rape is the explanans is not part of the evidence given for the theory in which rape is the explanandum, so disputing the first doesn’t undermine any of the reasons given for believing the second.

Again, speaking generally, you seem to be awfully muddled about causal claims, implication, and the proper places in which to attack an argument. This discussion is about causal claims, and causal claims are not claims about material implication. (A causal claim of the form “P’s being true causes Q to be true” is not truth-functional at all, because causal claims, among other things, have to support counterfactuals.) Maybe you are running into problems here because you believe that if someone advances a theory of the form “X causes R,” and another theory of the form “R causes P,” the direction of causation and the common middle term somehow suggests that the first theory is somehow logically prior to, a premise for, the second theory, and so that the evidential basis for the second theory somehow must depend on the evidential basis for the first theory. If you do believe that, I don’t know what to say except that it’s a hopeless muddle of really distinct causal, logical, and epistemological relationships, and you need to try to more carefully distinguish claims about complex causal chains between events from claims about complex logical and evidentiary relationships between statements asserting the existence of simple causal chains between events.

If that’s not what you’re confused about, then you’ll have to state more clearly why exactly you think a discussion of Brownmiller’s theory about the causes of rape has any evidential bearing on her theory about the effects of rape, and also just what precisely the antecedent is supposed to be and what the consequent is supposed to be in the material conditionals you keep trying to use.

  • Jerry: “your post contains quotes from women that basically blame men for most violence in the world (MacKinnon’s quote especially)”

MacKinnon’s quote does not say anything at all about either what absolute quantity or what proportion of the violence in the world is committed by men rather than women. What she actually says is that men commit some violence against other men (she doesn’t say how much), and men commit some violence against women (she doesn’t say how much), and then she contrasts the different ways in which the one kind of violence and the other are committed. I’m sure she has views on that, and so do I, but those views aren’t expressed in the quote and they aren’t material to this discussion. Any claim about how far men are to blame for how much violence is a claim that you have projected into the quote, not something that was there to be found.

  • Jerry: “Your claim that you are not saying men are bad and that it is just the science that makes them that way . . .

I literally have no idea what this means. I have not advanced any theory at all about what either the causes of rape are, or what the moral status of men, either individually or collectively, may be. I also have no idea what you mean by “the science [making] them that way.” What science? What claim are you even referring to?

  • Jerry: “You and Brownmiller have done none of that accounting.”

I already told you that I’m not attempting to provide a comprehensive defense of Brownmiller’s claims against all possible objections; if you want that, you should read Brownmiller’s book. My aims for a mid-length blog post are quite different. As for Brownmiller, unless you have read her book (I mean the whole thing, not just the handful of quotations that I or somebody else has pulled for brief consideration), then you have literally no idea at all what she does or does not account for.

  • Jerry: “Ad hominem covers the kind of insult you used, which dismisses the argument by demeaning the target of the insult as someone that unfairly shouts and worse, shouts irrelevancies.

Jerry, characterizing your argument as irrelevant is, I repeat, not an argumentum ad hominem. It is not an argumentum of any kind, because it has no internal inferential structure. It’s an assertion about your argument, which happens to be the conclusion of an argument drawn from a distinct set of premises. You might find the characterization, or the wording in which it is expressed, insulting. But “statements which you find insulting” and “examples of argumentum ad hominem” are two distinct classes, and their members have different logical properties.

As for that argument from distinct premises, I provided several reasons in my comments for saying that your reply was largely irrelevant to the point you were supposedly replying to. Those reasons may be good reasons, and they may be bad reasons, but they are reasons which had specifically to do with the structure and direction of the argument itself, not with any of your personal characteristics or circumstances as the person advancing the argument. You cannot simply point at the conclusion of an argument, declare “I find that conclusion insulting!” and then write off the entire argument as an exercise in argumentum ad hominem. (Or rather, you can’t do that without proving that you don’t understand what the term “ad hominem” means.) Argumentum ad hominem hasn’t anything to do with your reaction to the conclusion; it has to do with the kind of premises that the argument appeals to.

  • Jerry: “‘Since kharris disagrees with you, and DRR tells you to stop, I must consider I have won the argument!!! ZOMG!'”

I didn’t say that I won an argument. I said that you were devoting a lot of energy to topics that weren’t on-topic for the discussion, weren’t responsive to the specific claims advanced in my post, and which a number of people have repeatedly said they’re not much interested in discussing at length with you.

  • Jerry: “Here is a transcript of a speech from Wendy McElroy. . . .

I’m not interested in your views on male victims of domestic violence, or the ERA, or on the debate between liberal and radical feminists, or your beef with contemporary feminism broadly. These issues have nothing to do with the proper interpretation of Susan Brownmiller’s theory about the systemic effects of stranger-rape.

Re: A Spontaneous Order: Women and the Invisible Fist

  • Jerry: “P.S. Ad hominem attack: ‘You’ll have to engage with that if you want to actually join the conversation, rather than just shouting irrelevancies at it,’ — I think it’s clear I have not been shouting irrelevancies, regardless of how you would like to characterize my argument or me.”

That’s not an argumentum ad hominem. It is not even an argument at all; it’s a piece of advice which neither draws from premises nor moves towards a conclusion. It contains an implied characterization, which you may find personally insulting; but insults may be either called-for or uncalled-for, depending on the breaks, and are not the same thing as argumentum ad hominem, which is always a logical fallacy.

  • Jerry: [after a quotation from Brownmiller and a quotation from MacKinnon] “This is what RadGeek buttresses her argument with.”

No, it’s not. You seem to be having consistent problems with understanding the direction of inference in arguments. (For example, you also have repeatedly spoken as if the part of Brownmiller’s theory that was under discussion in the post was attempting to explain or make predictions about the causes of stranger rape. It’s not; it’s about the effects.) Here, you have failed to grasp that my post was intended to EXPLAIN THE CONTENT of the claims in those quotations using terms which a certain part of my audience would be likely to understand and find interesting.

The post was not intended to establish some further conclusion BY MEANS OF those quotations. The quotations are not introduced as evidence for a conclusion. They are introduced as texts to be interpreted; the evidence for the interpretation I favor is provided elsewhere in the post.

  • Jerry: “RadGeek’s point of departure is dubious and weak. Her conclusion seems to be . . .

My conclusion is that Susan Brownmiller is advancing a theory on which patriarchy is substantially reinforced by a spontaneous order arising from the effects of pervasive, random acts of sexual violence against women.

Any other suggestion as to what my conclusion “seems to be” is sure to be overreaching on your part.

As for your beefs with a random assortment of popular feminist bloggers, other claims that Susan Brownmiller happened to make about a different topic (e.g. false report rates), Women’s Studies programs in Universities, feminist analyses of domestic violence, social constructionism, zero tolerance policies, or the price of tea in China, I honestly don’t care. Judging from the response that your comments has gotten, I doubt much of anyone else here does, either. I’m sure that these issues are all very important to you, but they are not actually material to my post, or to the part of Brownmiller’s theory that’s under discussion, or to the discussion that basically anyone other than you has been pursuing. I would call them red herrings, but even an accusation of misdirection would require a degree of coherent direction that your posts have, so far, not demonstrated.

  • kharris: “By the way, RadGeek admits to Women’s Studies, but so far, not to being female that I can tell.”

For what it’s worth, I’ve only ever taken one course in my life that would qualify as a “Women’s Studies” course, and it was a fairly straightforward Psychology of Sexuality course, which wasn’t especially feminist in content. (It was cross-listed as Women’s Studies but taught by regular Psych department faculty.) Not that I think there’s anything wrong with taking Women’s Studies courses; that’s just not the way my academic career panned out.

However, I will happily concede just about any empty polemical label that jerry wants to throw at me, without argument, because I don’t give much of a damn what he calls me, and I’d just as soon get it out of the way in order to discuss something that matters.

She blinded me with science

How would you know how much or how little of “the research” on this topic I have read? The short answer is that you have no idea, because you haven’t begun to engage with the arguments at hand. If you want to demonstrate that there are important facts of which I am not aware you will have to, you know, actually point them out, and not just wave your hands at The Science and The Research without discussing any concrete findings.

Jonathan: For example, because the study of race was used in the past for nefarious purposes, discount any current data on race, even if it means potentially more effective medications for people of certain races.

This is a strawman. Nobody suggested that you should “discount any current data” on so-called “mental illness.” I said that the history of psychiatric abuse is a good reason for caution in appeals to consensus, which is something different. Similarly, people certainly should be cautious of research on racial difference, given the history of racialist pseudoscience. The epistemic authority of a scholarly community depends (in part) on its reliability in getting to the truth, and when there’s a long history of pseudoscience being promoted in order to provide ideological cover for prejudice, there is nothing logically askew in exhibiting a healthy degree of skepticism.

This is hardly ancient history. For example, Homosexuality was a recognized “disorder” until 1974, and so-called “Gender Identity Disorder” remains in the DSM to this day. Walter Freeman was still cutting up people’s brains not 40 years ago.

Of course, if you have an actual argument or empirical data to present, rather than just an appeal to The Science, then that argument and that data can and should be evaluated on its own merits, independently of whatever historical worries one may have. But since you have produced nothing of the sort, there is nothing to be assessed on its own merits; we have only the appeal to authority.

Jonathan: I’m not making an appeal to authority or scientific consensus.

Dismissing an argument on the grounds of an assertion, without further evidence, that its conclusion is “backwards, anti-science, and ignorant” is one of two things. Either it’s (1) an appeal to authority, which can sometimes be a cogent form of argument in the right context, or (2) simple abuse in place of an argument, which never is. I took the more charitable interpretation of supposing that you intended for your remarks to be (1), and so gave an argument as to why the appeal is misplaced, in this context.

If I misunderstood you, and should have adopted the less charitable interpretation, well, I guess I apologize.

Two points

Micha,

Two things.

  1. You write: “If people who suffer from mental illnesses believe (or are convinced by believing friends and family) that mental illness is a myth, they may not get the help they need, and suffer greatly as a result.”

But Szasz’s views, if implemented, would not deprive the people currently labeled “schizophrenic,” “bipolar,” “depressed,” etc. from “getting help” for their problems. The notion that it would is based on a common but extremely careless misreading of Szasz’s argument. Szasz is quite explicit that the symptoms on which a diagnosis of these pseudo-“diseases” are based are quite real, and pose real problems for people’s lives. What he objects to is the philosophical and political leap of organizing the understanding and “treatment” of those symptoms under pseudomedical terms like “schizophrenia,” “bipolar disorder,” “depression,” etc., which ultimately have a lot more to do with the legal requirement that someone have a diagnosed “disease” in order to get most drugs, than they do with any real scientific basis for the claim that all these symptoms trace to a single, identifiable disease.

In Szasz’s ideal society, people who are suffering from what is now mistakenly called “mental illness” would in fact have far more access to help than they have now, since the abolition of pharmaceutical laws and government licensure laws would remove a couple of the major barriers to entry and price floors on psychiatric “help.”

Szasz also believes that psychiatrists should not have the power to force “help” onto their “patients” against those patients’ wills. But the power to force “help” on unwilling “beneficiaries” is quite a different issue from the ability to “get help” when one needs it.

  1. You write: “But shouldn’t one of our goals be truth?”

Sure. But summary dismissal of an argument based on an unsubstantiated assertion that it is “backwards, anti-science, and ignorant” is not, as I see it, a reliable method of getting to the truth. Especially not when there are specific historical reasons to be cautious of “consensus” in the field in question, and when the person whose arguments are being thus dismissed without discussion is in fact a dissenter within the same community of experts whose authority is supposedly being referred to. (In this respect, Szasz’s position, as a professionally trained medical psychiatrist, is quite different from that of creationists who have no training in paleontology and evolutionary biology, or Holocaust deniers who have no training in history. That makes an appeal to authority, rather than an critical engagement with Szasz’s specific arguments, rickety in the extreme.)

Science!

Well, hell, if someone in a white lab coat says it, it must be true.

Certainly there is no reason to be cautious of appeals to scientific consensus in this field. There are certainly no prominent examples of appeals to scientific authority, or to psychiatry in particular, that had little to do with science and lots to do with providing cover for coercive normalization, inhuman “treatment” of so-called “patients,” or torture and brain damage posing as “cures.” No need to think for yourself, citizens; the experts have already done it for you.

Meanwhile, we can just safely ignore Szasz’s actual arguments, ignore the extent to which common criticism and “embarassment” over Szasz is based on easily refuted strawmen (such as the claim that he believes that hallucinations are “made up”), and substitute an appeal to authority and an ad hominem (abusive form).