This argument of Peikoff's strikes me as quite wrongheaded. If a large percentage of people who smoke get lung cancer, then there is good reason to believe that smoking causes cancer. There are a great many drugs whose mechanisms are unknown.

Incidentally, Peikoff's comment tends to confirm what Barbara Branden says in PAR about Rand's smoking.

WSS,

I hardly ever visit Leonard Peikoff's website, but happened to spot this one around the same time you did.

In the early 1970s, when Dr. Peikoff (and Ayn Rand) were still smoking vigorously, he inveighed against reliance on statistical correlations. I remember the issue coming up in one of his history of philosophy lectures, and (in more detail) in his lectures on logic. He specifically targeted conclusions about the effects of smoking on health. I thought it was BS back then--so did most of the other folks who were listening to the tapes with me.

Peikoff makes it sound like he and Rand quit smoking cigarettes because some biochemist took them aside and, with the aid of a whiteboard, explained how burning leaves give off aromatic compounds like benzopyrene, which is close enough in shape to a couple of the DNA bases to be a powerful mutagen, etc. etc. etc.

What happened in 1975 is that Rand got a chest X-ray. Her doctor pointed a cloudy area on it, and told her she had to quit smoking ASAP. She promptly underwent an operation in which part of one of her lungs was removed. She survived the lung cancer, albeit in a significantly weakened condition.

At least, that's what The Passion of Ayn Rand says happened. I didn't know the whole story in 1975, but I recall there was a long hiatus in the production of the Ayn Rand Letter, which Rand eventually explained by saying she was recovering from a major operation. The news gradually traveled through the grapevine that the operation had been for lung cancer.

Robert Campbell

This argument of Peikoff's strikes me as quite wrongheaded. If a large percentage of people who smoke get lung cancer, then there is good reason to believe that smoking causes cancer. There are a great many drugs whose mechanisms are unknown.

Incidentally, Peikoff's comment tends to confirm what Barbara Branden says in PAR about Rand's smoking.

Thanks for the note, Neil. What struck me was this, "Statistics, she held, may offer a lead to further inquiry but, by themselves, they are an expression of ignorance, not a form of knowledge. For a long period of time, as an example, there was a high statistical correlation between the number of semicolons on the front page of The New York Times and the number of deaths among widows in a certain part of India¸"

Larry Budd moment, only if this is all truly believed by Larry Budd. Otherwise, LP is off the hook.

So, where did he get this "[Ayn Rand held that statistics] may offer a lead to further inquiry but, by themselves, they are an expression of ignorance, not a form of knowledge."?

Is this true? If so, where did she hold forth on the topic? That would make interesting reading today in light of squabbles over climate change.

Is this true? If so, where did she hold forth on the topic? That would make interesting reading today in light of squabbles over climate change.

Our inimitable Robert Campbell has set the stage upscreen. Thanks! But can we find the Stats == Ignorance cite without torturing Larry Budd?

Robert,

That's an excellent point. (BTW, when I said that Peikoff's version tended to support Barbara Branden's account in PAR, that I was referring to Rand being dismissive of the statistical correlation between smoking and cancer.)

Incidentally, if Barbara'a account is correct (and I don't have any reason to doubt it), it's pretty clear that when Peikoff recounts some aspect of Rand's life or personality (that she had a temper), we are entitled to ask what he is omitting.

See this thread in the Epistemology forum:

http://www.objectivistliving.com/forums/in...?showtopic=2558

And in particular this post of mine:

http://www.objectivistliving.com/forums/in...ost&p=16972

I signed on specifically to respond to Shayne's query, which I'd noticed earlier, and to say (I wrote this before I signed on):

"Whoever told you that is right, Shayne, although the expression should be 'contextually certain.' Peikoff waxed at length on exactly that example, on which Rand (with her dislike and suspicion of statistics) had waxed at length to him. I haven't time for providing details. Maybe one of the other 'old-timers' here will notice your post and respond."

Peter Reidy has meanwhile posted to fill in some details. Rand was extremely sceptical of statistics as an epistemological method. (This scepticism, btw, was a significant factor in her not placing any credence in the early smoking-and-cancer studies. Another factor was that she thought cancer was caused by "bad premises.")

Ellen

___

I'm wrong, however, in something I said there. The expression Peikoff used WAS metaphysical vs. epistemological possibility. "Contextual certainty" is related but isn't identical. I'm not sure when the idea of "contextual certainty" started being talked about.

Rand had trouble "making sense" of HER having gotten cancer, because of her belief that cancer was caused by "bad premises." I think this is discussed somewhere else in the forum, amongst the early threads, but I don't remember where.

I'll add that actually she was right not to place credence in the early cancer studies, although not for the reasons she gave of generalized distrust of statistical methods -- instead because there was jumping the gun in the Surgeon General's Report. Long story there (including that the dog studies were found to have had a systematic error in the reading of the slides). I worked for about a year and a half in a couple different capacities at the headquarters of the ACS and I heard the low-down about what really was known and what wasn't when Koop issued his report. However, Rand wouldn't have had any way of being privy to this information. Her objections were (a) general objection to statistics; (b ) specific belief in the bad-premise etiology of cancer.

Ellen

___

WSS,

I don't think we can improve on Leonard Peikoff's testimony that Ayn Rand equated statistics with ignorance.

The reason is that Rand published virtually nothing on the subject. I don't recall even posthumously published references to statistical methods (say, in her journals, or in the edited transcripts from her epistemology workshops). In ITOE, there's a poke at psychologists who imagine they can measure "the human psyche" using "statistical questionnaires," but it's not obviously directed at correlational methods.

The March 1964 issue of The Objectivist Newsletter ran a review by Joan Mitchell Blumenthal of a book titled The Tyranny of Testing. Blumenthal agreed with the author's critique of multiple-choice tests (such as the SAT) and the manner in which "validity" is determined for questions on these tests. Again, though, the statistical procedure that Blumenthal was complaining about is part of "psychometrics."

It's worth noting that nearly everything that Dr. Peikoff said about statistics in his lectures on logic was standard-issue stuff, e.g., not being able to infer causation from correlational data alone. It was the specific comments on smoking and lung cancer that suggested a deeper objection.

Robert Campbell

PS. Perhaps Ba'al would have a different take on the matter, but I can't approve of torturing Larry Budd to get this information