Good Math, Bad Math

I recall Ronald Fisher himself argued against smoking as a cause of lung cancer. At the time I think he argued they were misusing his methods and the results did not support the conclusion, but that is not exactly the same as saying smoking does not cause lung cancer.

Posted by: George Larson | July 19, 2007 1:48 PM

I think you might be opening yourself for attack by not posting volumes of the carcinogens in cigarettes and whether or not all of those are released into the air. For example, benzo(a)pyrine may take far less to cause cancer, but how much of it is released into the air. While its potency may be orders of magnitude higher than alcohol, its volume may be orders of magnitude lower.

Disclaimer: I'm just filling a hole in the argument, not taking the other side.

Posted by: Brent | July 19, 2007 2:55 PM

Bravo, Mark!

I actually was going to take on that bit of craziness myself. I'm glad you made it so I didn't have to. By the way, McFadden is trying to defend his nonsense back on my blog.

Posted by: Orac | July 19, 2007 3:09 PM

I nominate Mr. Douchebag for the Fields. That's... simply brilliant.

Posted by: Xanthir, FCD | July 19, 2007 5:54 PM

Heh... finding this trail of blogginess makes me think back to the comment (by Mark H? Orac?) yesterday or so about the "Crank Bat Signal" summons.

Mark C., no dishonesty involved at all. I was unaware of the existence of the tables and information you post above and will examine it. At the moment though I haven't even gotten back to the denialism/insolence blogs where I believe I am both awaiting some responses and owe a few.

I noticed your selection of carcinogens though and am wondering, since you seem to be aware of such things, why you did not choose the proper carcinogens for comparison: i.e. the unique Class A carcinogens in tobacco smoke and their quantities. See the 10th IARC report at http://ehp.niehs.nih.gov/roc/toc10.html#search (I believe there's an 11th by this point but don't have the URL handy) and the SGR '89 or '06 for particulars.

I think if you do so, and THEN apply your analysis, you may come up with a somewhat different conclusion (you may not also... I haven't done that analysis myself).

I also wonder why you seem to think that tobacco smoke in the air would behave so much differently than alcohol vapors. Granted one is a gas form and one a particulate, but I believe that very small particulate matter behaves aerodynamically in ways quite similar to gas. Do you have evidence or reason to believe to the contrary?

And a final note: I believe your and your blog friends are consistently missing the MAIN point I was making: it *IS* crazy to worry about such exposures to alcohol, and it is similarly crazy to worry about such exposures to secondary smoke. You can argue for a degree of difference, even a *great* degree of difference, but as the Surgeon General has so painstakingly pointed out, "there is no safe level of exposure" to carcinogens. So was he lying? Or is it dangerous to be near you while you're sipping your Beefeater?

Or *is* there perhaps a "safe level" in normal parlance that is simply ignored and twisted by the antismoking lobby?

Michael J. McFadden
Author of "Dissecting Antismokers' Brains"
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 19, 2007 9:21 PM

Evidently my initial posting either didn't take or I posted twice to another in this mess o' blogs. Mark, here you are...

======

Heh... finding this trail of blogginess makes me think back to the comment (by Mark H? Orac?) yesterday or so about the "Crank Bat Signal" summons.

Mark C., no dishonesty involved at all. I was unaware of the existence of the tables and information you post above and will examine it. At the moment though I haven't even gotten back to the denialism/insolence blogs where I believe I am both awaiting some responses and owe a few.

I noticed your selection of carcinogens though and am wondering, since you seem to be aware of such things, why you did not choose the proper carcinogens for comparison: i.e. the unique Class A carcinogens in tobacco smoke and their quantities. See the 10th IARC report at http://ehp.niehs.nih.gov/roc/toc10.html#search (I believe there's an 11th by this point but don't have the URL handy) and the SGR '89 or '06 for particulars.

I think if you do so, and THEN apply your analysis, you may come up with a somewhat different conclusion (you may not also... I haven't done that analysis myself).

I also wonder why you seem to think that tobacco smoke in the air would behave so much differently than alcohol vapors. Granted one is a gas form and one a particulate, but I believe that very small particulate matter behaves aerodynamically in ways quite similar to gas. Do you have evidence or reason to believe to the contrary?

And a final note: I believe your and your blog friends are consistently missing the MAIN point I was making: it *IS* crazy to worry about such exposures to alcohol, and it is similarly crazy to worry about such exposures to secondary smoke. You can argue for a degree of difference, even a *great* degree of difference, but as the Surgeon General has so painstakingly pointed out, "there is no safe level of exposure" to carcinogens. So was he lying? Or is it dangerous to be near you while you're sipping your Beefeater?

Or *is* there perhaps a "safe level" in normal parlance that is simply ignored and twisted by the antismoking lobby?

Michael J. McFadden
Author of "Dissecting Antismokers' Brains"
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 20, 2007 3:11 AM

Speaking of equivalent-time dosages, secondhand smoking is perhaps more dangerous than what the filter-protected smoker ingests.

In any case, IIRC there have been court cases here where persons have gotten compensation for lung cancers that can be directly traced to their work in former smoking environments. So it seems the casual connection can be fairly ascertained besides epidemic studies.

The most damaging to McFaddens argument is IMO that alcohol is secondarily carcinogenic and AFAIK most airborne light alcohols are vented efficiently by the lungs before they get metabolized.

Smoke particles OTOH are vented within 15 minutes. (Proposed time limit between smoking and entry to clean rooms because at that time excess particles become immeasurable compared to non-smokers.) But my guess is they mostly stick and/or in any case leave much of the tar, or we wouldn't see smokers get lung cancers.

And the smoke particles, often highly carcinogenic as such, by this venting often ends up in other persons lungs. Smoking is a gift that goes on giving.

Brent:

I think you might be opening yourself for attack by not posting volumes of the carcinogens in cigarettes and whether or not all of those are released into the air.

Those numbers can be extracted from McFadden's claims in the quote. But I don't think the 60 % benzene makes it much better. The same site gives benzene carcinogenic TD50 potencies of 169 (77.5) mg/kg/day for rat (mouse). Sure, you can calculate another mean. But the caveat that tar and alcohols differ remains.

Posted by: Torbjörn Larsson, OM | July 20, 2007 8:16 AM

At that rate of evaporation, a typical UK measure of spririts (35ml, 40% by vol) would contain no alcohol whatsoever after approximately 14 hours. I find that extremely difficult to believe. Leave a whisky out overnight, and it's still whisky in the morning - I can confidently assert this from personal experience.

I've just put 36 grams of gin in a beaker in my study. (No-one in the house drinks it and we don't have any cats.) I'll keep you posted.

Posted by: Stephen | July 20, 2007 10:45 AM

I will try to answer folks briefly and to the point here:

Anonymous wrote, " mr. mcfadden, i'm curious. how exactly does the funds available to a certain party and their potential level of 'propaganda' have any bearing on whether the message is right or wrong? ...

(It doesn't.)

oh and further, for all we know, 800+ million dollars is petty change in the business. incomplete math for the win.

(Please drop 1/1,000th of your next petty change drop off at my door. Thank you.)

T. Larsson wrote, "Speaking of equivalent-time dosages, secondhand smoking is perhaps more dangerous than what the filter-protected smoker ingests."

("Speaking of equivalent-concentration dosages, secondhand smoking is probably about 1/1,000th as dangerous as what the filter-protected smoker ingests... or less.")

In any case, IIRC there have been court cases here where persons have gotten compensation for lung cancers that can be directly traced to their work in former smoking environments. So it seems the casual connection can be fairly ascertained besides epidemic studies.

(The biggest case in this area was the Florida flight stewardesses case. That was settled, not decided, for $300 million of which nothing at all went to the "persons". If my memory serves me correctly I believe Florida law was rewritten specifically in order to make it impossible for Big Tobacco to walk away on that one. In any event, science usually does not establish "causal connections" on the basis of a ruling by one or two judges or juries or such... unless your name is Galileo.)

The most damaging to McFaddens argument is IMO that alcohol is secondarily carcinogenic and AFAIK most airborne light alcohols are vented efficiently by the lungs before they get metabolized.

(I'm not sure what you mean by "vented efficiently"? Can you supply a reference on that?)

Smoke particles OTOH are vented within 15 minutes. (Proposed time limit between smoking and entry to clean rooms because at that time excess particles become immeasurable compared to non-smokers.) But my guess is they mostly stick and/or in any case leave much of the tar, or we wouldn't see smokers get lung cancers.

(I'm not clear on the "vented efficiently" by the lungs statement and how it relates to "clean rooms". I would think that air changes per hour would be a more important measure than a simple time element.)

And the smoke particles, often highly carcinogenic as such, by this venting often ends up in other persons lungs. Smoking is a gift that goes on giving.

(Some of them do obviously. Under decent ventilation conditions probably less than 1/1,000th. I don't know if that qualifies as "often".)

Dunc quoted someone:

#16we'll assume that McFadden is being entirely honest in his estimate that a martini sitting on a bar releases one full gram of alchohol into the atmosphere per hour
And then wrote, "Lets not, huh? At that rate of evaporation, a typical UK measure of spririts (35ml, 40% by vol) would contain no alcohol whatsoever after approximately 14 hours. I find that extremely difficult to believe. Leave a whisky out overnight, and it's still whisky in the morning - I can confidently assert this from personal experience."

(Dunc, you or anyone else here is MORE than welcome to replicate the very simple experiment outlined in my BMJ response. Given the feelings and competence exhibited by many of the posters here I'd be quite surprised if no one has already tried it. And I'm quite sure that if someone DID try it and found I was being dishonest we would have heard about it louder than Gabriel's call. Remember the conditions: standard ventilation/temperature room, martini glass, 48g (1 large shot) of grain alcohol (95% Everclear will do fine, 100 proof vodka MAY still have some largely water remnant at the end of 48 hours... give it a try if you want.)

(And, re an earlier post about 10 minutes per smoke vs. an hour for my martini cutting the 2,000 times down to 330: that holds true as a comparison ONLY if you gulp your martini in 10 minutes and then sit there with an empty glass and hold your breath and don't perspire for the next 50: remember, once that alcohol goes into your body your body starts getting rid of it too. The excretory addition to the equation is something I didn't bother bringing up in the original model since I have neither the funds nor the expertise to measure it.)

Stephen quotes and notes:

#17At that rate of evaporation, a typical UK measure of spririts (35ml, 40% by vol) would contain no alcohol whatsoever after approximately 14 hours. I find that extremely difficult to believe. Leave a whisky out overnight, and it's still whisky in the morning - I can confidently assert this from personal experience.
I've just put 36 grams of gin in a beaker in my study. (No-one in the house drinks it and we don't have any cats.) I'll keep you posted.

(When I was staying in England I noticed pubs seemed to sometimes offer 35ml and sometimes 50ml in their standardized pours. Remember though: unless you've got the money/equipment for some fancy measuring you won't be able to determine the percentage of alcohol remaining: that's why I recommend grain for the experiment. Of course if the gin DOES completely evaporate even though it's half composed of less volatile water, then my point is doubly made.)

#18

Mark C wrote, "Michael McFadden: So what you're saying is:
* You're a serious researcher of the link between smoke and cancer.

(No. I am a serious, though completely unfunded and unaffiliated researcher, of the relationship between secondary smoke exposure and cancer/heartdisease etc. I am also a researcher concerned with the social movement surrounding and promoted by the perception of that relationship.)

* You've submitted scientific papers on the matter, which were rejected
entirely for political reasons.

(No. I've submitted one scientific paper on the matter, which I believe has so far been rejected largely for political reasons. I wouldn't be presumptuous enough to say "entirely" as I am clearly not well-experienced in the area of publication of research papers and it is possible our paper could reasonably be rejected on non-political grounds. I believe the opposite was more the case here though.)

and
* You're completely unfamiliar with the idea that the degree of
carcinogenicity varies for different substances.

(No. I am not completely unfamiliar with it. I do not claim to be a full-time and highly-trained medical expert however and was not familiar with the comparative figures/table you brought up. You evidently are however, and I am awaiting your findings regarding the 7 recognized Class A Carcinogens in secondary smoke. I do not rule out that your contention might be valid, but I'd like to see the evidence.)

Do you really expect anyone to take you seriously when you
say things like that? If you're not even familiar with something as
totally trivial as dose/response variation,

(er... do you mean like Surgeon General Carmona's "there is no safe level" ? Or the claims about 30 minute heart attacks and vasodilation based upon exposure levels ranging from 300% to 2,000% the average in the middle of smoking sections on pressurized aircraft? Seriously, I think I have made quite clear in my writings that I am aware of and concerned about dose/response variation. If you enter you'll get 21 hits.)

then it's pretty much impossible
that you've written a remotely credibly scientific analysis of the health impacts of second-hand smoke.

(If you find any defects with my analyses please state what they are. You have indeed pointed out one possible defect and I'm awaiting a fuller statement of it. **HOWEVER** it is important to remember the core of what has brought us here today: the fact that the Helena/Pueblo studies did NOT examine the effects of smoking bans on nonsmokers and yet have VERY consistently and willfully represented by the authors and by responsible authorities as having done so; and the fact that the journals responsible for the initial publications of both have failed to offer a reasonable podium for a responsible scientific countering of their earlier misguided publications.)

The key question about
second hand smoke is whether the quantity of dangerous substances in
cigarette smoke inhaled by a non-smoker is sufficient to cause harm. If you don't even understand the idea of different quantities of different substances having different degrees of effect, you simple can't have done
the analysis - because you show no comprehension of *the* most fundamental thing that you'd need to analyze.

(See above note re: "the dose makes the poison.")

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 20, 2007 2:56 PM

I gather that any further figures are largely superfluous for the discussion. However after 5 hours there is as yet no measurable evaporation from my gin (i.e. a gram at most), whereas according to McFadden 5 grams of alcohol plus an unknown amount of water should have evaporated. I'll keep you posted.

Posted by: Stephen | July 20, 2007 4:19 PM

Mark, hmm... usually if I try to refute someone's argument I provide the evidence for that refutation rather than simply say, "Hey, you should go check these other things yourself!" I wonder how far I would have gotten with the Helena authors using that approach? Doubt if it would have been very far considering they won't even defend their work and statements when evidence IS provided (even if that evidence had to be dregged from the WayBack Machine after they deleted it).

I will look over your table more closely later today if I can. A preliminary examination shows that it seems to be limited to rat/mouse carcinogenicity which is suggestive of human effect but generally not considered as much more than suggestive (which is one of the reasons why we have the different "Classes" of carcinogenicity). Even within such closely related species as rats and mice the table shows huge variations with Benzidine seeming to be ten times as carcinogenic in mice as in rats and Benzofuran being almost twenty times as carcinogenic in rats as in mice.

I still feel that the proper burden lies upon you Mark: you clearly "started" the attack upon my work in general, and if you feel you have found a weakness or oversight you should be ready and able to demonstrate the results of that weakness/oversight.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Anonymous | July 20, 2007 5:07 PM

As for Galileo, the shoe is on the other foot: I was saying that taking a court ruling to be meaningful as scientific evidence (as you seemed to be doing) was not valid.

The only reason to mention Galileo here is because you feel persecuted by court (in lieu of the Inquisition). Or, at worst, I am misreading you as you were misreading me.

Posted by: Torbjörn Larsson, OM | July 20, 2007 7:04 PM

TL wrote, "The only reason to mention Galileo here is because you feel persecuted by court (in lieu of the Inquisition). Or, at worst, I am misreading you as you were misreading me."

I'm sorry Torbjorn. I think our language difference may have played a part in the misunderstanding. You'd written:

"...there have been court cases here where persons have gotten compensation for lung cancers that can be directly traced to their work in former smoking environments. So it seems the casual connection can be fairly ascertained besides epidemic studies."

And I was saying in response that a court ruling does not determine scientific fact. That's why I brought up Galileo: it's the classic example of a political position overruling science.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 20, 2007 10:01 PM

Alchohol analogies aside, I've always considered the claim that "second hand smoke is as dangerous or even more dangerous as is first hand smoke" to be obviously deceptive slant. After all, the second hand smoker is inhaling drastically diffused smoke compared to the first-hander, even when the first hander is using a filter (the 1000x less figure seems flexible, since the precise difference would obviously be altered by specific, significant variances in room ventilation, i.e., windows, air conditioning, etc.)

All the above debate doesn't change this obvious point, so the heart of McFadden's theme (that second hand smoke is radically less dangerous as typically inhaled than is first hand smoke) appears correct, in spite of the nitpicks re the details.

Posted by: Norm Breyfogle | July 21, 2007 1:49 AM

McFadden:

And I was saying in response that a court ruling does not determine scientific fact. That's why I brought up Galileo: it's the classic example of a political position overruling science.

I understood that, obviously. But the subtext was that it was a ruling that you didn't agree with. And it is also the classic example that crackpots put forward to defend their position.

Now, normally one would choose the lesser of two cigarettes when reading a text. But the context here favored my reading.

Also, I'm curious. What do language difference do you see?

I see that you have severe difficulties to read my comments, but it seems to be more on the logical level. For example, I discuss tobacco smoke particles venting from lungs in the context of measurements for clean room technology, and you jump to the unwarranted conclusion that I discuss particles venting from clean rooms.

That, and the fact that you are nitpicking as in this comment, made me confine my previous answers to a subset of your questions. If you feel that it was a language problem on any of our parts, perhaps I should reconsider that.

Posted by: Torbjörn Larsson, OM | July 21, 2007 7:48 AM

All the above debate doesn't change this obvious point, so the heart of McFadden's theme (that second hand smoke is radically less dangerous as typically inhaled than is first hand smoke) appears correct, in spite of the nitpicks re the details.

Agreed, and as I believe introduced that slant, it goes to potentiality as you say "typically inhaled". It isn't often that second-hand smokers sits in such a totally smoke-saturated environment where the filter smoker gets cleaner air. But IIRC it can happen.

So we would have to take steps to avoid it - prohibiting smoking in public spaces seems like an obvious act.

Posted by: Torbjörn Larsson, OM | July 21, 2007 8:04 AM

"Agreed, and as I believe introduced that slant, it goes to potentiality as you say "typically inhaled". It isn't often that second-hand smokers sits in such a totally smoke-saturated environment where the filter smoker gets cleaner air. But IIRC it can happen.

So we would have to take steps to avoid it - prohibiting smoking in public spaces seems like an obvious act."

Interesting jump. Which brings us back to the point. Which is -- that such zero tolerance measures are unnecessary, and not in line with the treatment of other things in society that people enjoy and/or find necessary or convenient which may even be potential hazards. However you all work out the math, I am assuming from Marc C.'s figures that there is SOME level of carcinogenic alcohol in my local bar. (Or I can say, as some smoking ban proponents, "If I can smell it, it must be there") Am I concerned?, not one bit. I believe Mr. McFadden's original point was a social choice between zero tolerance and other sorts of accomodation.

Posted by: GDF | July 21, 2007 11:52 AM

OK, my gin has now been standing for 26 hours, and we finally have some sort of result: an evaporation of 2.5g +/- 1g. Even if you generously assume that this is entirely alcohol, Dunc's skepticism was well justified. The statement "A standard martini releases roughly one full gram of the Class A carcinogen ethyl alcohol into the air in the space of an hour" is, as Jamie and Adam like to say, busted.

Posted by: Stephen | July 21, 2007 1:18 PM

Check the 1986 SGR and you'll find the studies cited there indicate exposures of anywhere from 1/100th of a cigarette per day to 2/10ths of one per day.

So you've already gone up from one 1000th (0.1%) to 1-20%. And when I looked up the 1986 SGR, the equivalent exposure figure cited was 0.1 to 1 cigarette per day. But what's a few orders of magnitude amoung friends?

And what is this "no safe limit" business you keep harping on? Are you under the misimpression that all substances with no safe limit are equally potent?

Posted by: trrll | July 21, 2007 4:10 PM

There have been a number of challenges/questions posed to me. I'll try to hit them all:

Stephen, I am surprised at your results. The 76% humidity may have something to do with it: I doubt the humidity levels in my house in late April or 2005 were anywhere near that. Humidity is fairly high in Philadelphia at the moment but unfortunately so is the temperature. I will redo the experiment using a martini glass with 48g of 100 proof vodka. If anyone else here would also like to share the experiment they would be welcome to.

TL, Let's accept that we just had an innocent misunderstanding about the Galileo reference. :) You say, "It isn't often that second-hand smokers sits in such a totally smoke-saturated environment where the filter smoker gets cleaner air. But IIRC it can happen." Can you provide a few examples? It shouldn't be hard to figure out the concentration of the smoke inhaled by the smoker using the standard 10 puffs at 20 ml per puff for total mainstream inhalation vs whatever the sidestream emissions are in an experimental situation comparable to my "standard small neighborhood bar" under the ETS Exposure button on my www.Antibrains.com website.

I'd recommend using either total "tar" or total "nicotine" for your measurements as picking just about any other component leaves the door WIDE open for distortion through cherry picking.

Now... speaking of cherry picking or distortion, Mark C. wants me to defend against his charge about the relative carcinogenicity of alcohol vapors compared to the Class A carcinogens in tobacco smoke. I believe there are six recognized Class A components of ETS: Arsenic, benzene, naphthylamine, aminobi-phenyl, vinyl chloride and chromium.

According to the 1999 Massachusett's benchmark tests, I believe a standard cigarette emitted on the order of 300 micrograms of benzene and less than 100 nanograms of the other five combined. I'll refrain from cherry picking the lit'luns and head straight for benzene.

As Mark pointed out the figure for alcohol producing cancer in 50% of rats was about 9,110 mg/kg/day. The figure for benzene was 169 mg/kg/day. Benzene then, at least for rats in this table, appears to be 54 times as carcinogenic as alcohol. So (just to keep it simple) let's say the true evaporation figure under standard temp/humidity conditions is actually .911g/hour. Then to be equal in "carcinogenic threat" we would need for a cigarette to emit .0169 grams... i.e. 1/54th the amount of the alcohol emitted from the drink.

Am I correct so far?

OK... as we saw above a cigarette actually emits .0003 grams. So we would need to burn 56 cigarettes to put out the same "total human carcinogenicity" in terms of benzene as our martini does. If we are willing to say that the average smoker will smoke two cigarettes in a bar/restaurant in the course of an hour, then it would take 28 smokers to be as deadly a threat to their companions as a single drinker.

Mark, if you would like to do the figuring for any of the other 5 Class A components of tobacco smoke you are welcome to. Since they are all present only in quantities of NANOgrams I kind of doubt you'll get results that would support your main contention that my comparison is totally without foundation.

HOWEVER... I will freely admit that my figure of 1 martini equaling 2,000 cigarettes should indeed have been corrected for this relativity and that it might be more correct to say that 1 martini = only about 50 cigarettes in terms of the threat to others.

Mark et al, if there are any problems with anything I have figured above, I am quite sure you will let me know...

And of course, a good bit of this argument rests upon my now disputed evaporation experiment. Do we have any profound neutrals here so far who'd like to give it a try? And, of interest, any chemists who'd like to discuss how the proportions of evaporation work for mixtures as opposed to straight grain alcohol? And how humidity would affect alcohol's evaporation speed?

::sigh:: Guess I'm gonna have to waste a good 48g of 100 proof vodka here on my own. Plus I gotta lock up that damn tipsy kitty!

:>
Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 22, 2007 2:33 AM

OK, the weather forecast for Philadelphia for the next two days calls for temperatures of 65 to 80 degrees with humidity levels ranging from 35 to 90%. I can't promise not to hit the air conditioner at some point in there, but any reduction in humidity should be compensated to some extent by a reduction in heat.

This time I have set up two martini glasses - very nice cobalt blue ones that I normally keep safe and in a china cabinet so there damn well better not be any earthquakes while they're sitting on top of a turned-off & unplugged TV - with one of them having about 48 g of water and one with about 48 g of 100 proof vodka. We'll see what happens. My hypothesis is that there will be significantly less evaporation of water than vodka and that at the end of 48 hours (3am EDT Tuesday morning) I'll have somewhere around 10 to 15g of water in one and about 0 to 5g of watered vodka in the other. But ... we'll see.

In terms of the SGR exposures: note that the figures you quote were, I believe, concerning shifts of 8 to 10 hours. They also concerned settings with air exchange rates that were probably 1/2 to 1/10th the air exchange rates that would be found in any decently ventilated modern bar or restaurant that was allowing smoking. Try recomputing your findings on the basis of a single hour and with those air exchange corrections and I think you'll find my estimates fall *well* within an order of magnitude.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 22, 2007 3:03 AM

Fundamentally, this is a "but Johnny's mother down the street lets him stay up late" argument. Most people learn by about age 10 that this is generally fruitless, because there the usual response is "well if he were my son, he'd have to go to bed earlier, too." Regulation of alcohol has at least as much to do with politics and practicality as science, and there may indeed be a scientific case for more stringent regulation of alcohol vapor. I can tell you that if we had levels of alcohol vapor in our lab comparable to a typical bar, we'd be in serious trouble with OSHA. but an that doesn't help you much with secondhand smoke.

However, from a biological point of view it is very bad science to base an argument on extrapolations based on carcinogenic potencies of a particular component of tobacco smoke vs. alcohol, extrapolated down to much lower concentrations than those used to study carcinogenic potency. The dose-response curves may be nonlinear, in which case a linear extrapolation will yield erroneous results. Moreover, tobacco smoke is not a pure chemical, but contains over a hundred toxic or carcinogenic compounds (with levels that are differ in sidestream smoke as compared to mainstream smoke) that may interact in the body to enhance or inhibit one another's effects. So while the sort of extrapolation McFadden is trying to do might be sufficient to obtain a preliminary estimate of risk, it could easily be wrong by orders of magnitude. Lacking epidemiological or experimental data to support it, it will carry very little weight with scientists. And obsessing at length about exactly how to do such an inherently unreliable calculation makes it appear that he does not really understand the science. While it is not exactly a scientific "howler" such as ignoring differences in carcinogenic potency, such fallacious reasoning is definitely the sort of thing that will negatively impact the chance that a paper will be accepted for publication in a good journal.

Posted by: Anonymous | July 22, 2007 1:00 PM

"Seems to me that an effective campaign against second hand smoke could be made without the use of such hysteria."

That's exactly the point. I respectfully disagree Norm, else, as I see it, the hysteria would not have been necessary. But the problem is we can't know because hysteria and ideology and political and financial interest are being repackaged now as science. (IMHO) And those who try to oppose this trend/tactic are dismissed as "cranks"

Posted by: GDF | July 22, 2007 1:29 PM

A smoker clearly inhales SHS while smoking in a bar or other communal smoking venue, but probably not as much as somebody who works there all day. But in any case, the epidemiology argues that the risks of SHS are less than those of active smoking--just greater than is considered acceptable in a workplace.

Posted by: trrll | July 22, 2007 3:23 PM

Anonymous wrote, " And obsessing at length about exactly how to do such an inherently unreliable calculation makes it appear that he does not really understand the science."

Er, Anonymous, *I* wasn't the one who was "obsessing at length" about it. My initial posting of the idea was meant to be a thought provoking piece in the BMJ. The blogger HERE was the one who began trying to analyze it down into the subcomponents and then challenged me to defend myself.

Anonymous also wrote, " While it is not exactly a scientific "howler" such as ignoring differences in carcinogenic potency, such fallacious reasoning is definitely the sort of thing that will negatively impact the chance that a paper will be accepted for publication in a good journal."

The paper in question has NOTHING to do with this subtopic of alcohol fumes other than that both discussions related to secondary smoke and the concept that worrying about one isn't much more rational than worrying about the other.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 22, 2007 7:06 PM

Norm:

McFadden is correct in complaining about second-hand smoke hysteria, especially when I've often heard (without reasonable qualifiers) that "second hand smoke is as bad or even worse than first hand smoke."

I don't agree with this being hysteria, since we are discussing epidemiological risks for bystanders. We could also add economical losses and losses in individuals freedom of choice which comes from creating lower quality environments, but that is another debate IMO.

Posted by: Torbjörn Larsson, OM | July 23, 2007 1:12 PM

Mr. Larsson,

Just curious about how you stand on a few things...

Do you believe that no worker should ever be allowed to work around ANY level of SHS (for example, a pool hall employee in a well ventilated hall who signs an informed consent?)

Should the law allow for a bar owner (let's say he's a smoker himself) who runs his own establishment, which is clearly posted on the door "smoking allowed"?

Should well ventilated "smoking rooms" in office buildings be illegal?

Should "cigar bars" be prohibited?

Should smoking be allowed in private clubs where the members prefer to allow smoking and one member is paid to tend bar? What if the club is a "smoking club"?

Posted by: GDF | July 23, 2007 2:04 PM

Perhaps I should have mentioned that all of the above are prohibited where I live. My second example is based on a real situation (although I don't know if he is a smoker himself) -- the guy is a Vietnam Veteran who opened a little "corner" tavern. It made the news because he replaced the American flag he used to fly with a Soviet one (as a symbol of protest).

Posted by: GDF | July 23, 2007 2:39 PM

Trrl, I myself *think* that only crazy people should be worrying about the effects of alcohol vapor in bars and restaurants, but on the other hand I freely admit to having done no research about it. If you yourself worry about it then I would suggest either staying out of those places or working to have alcohol service and consumption banned in them.

You are the one trying to make the case, not I. If you are comparing SHS smoke to alcohol vapor in an effort to show that SHS is benign, then the burden of proof is on you both to show both that alcohol vapor is indeed benign and that the comparison is valid. Suggesting that anybody who is not willing to share your unsupported assumptions must be "crazy" is rhetoric, not rational debate.

Posted by: trrll | July 23, 2007 5:53 PM

Trrl wrote, "If you are comparing SHS smoke to alcohol vapor in an effort to show that SHS is benign, then the burden of proof is on you both to show both that alcohol vapor is indeed benign and that the comparison is valid. "

The comparison was in indeed meant to indicate that the level of threat from both might be fairly equal, or that the threat from alcohol vapor might be much more. I personally think that the levels of such threats are benign. I believe I've shown the general validity of the comparison although I have not had my post of about 40 hours ago acknowledged by Mark C. yet.

I'm also still waiting for Mark C. to get back to me as to what level of exposure to carcinogenic alcohol fumes has been determined to be safe, and if none has, whether he believes drinking should be banned from workplaces.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 23, 2007 8:26 PM

The comparison was in indeed meant to indicate that the level of threat from both might be fairly equal, or that the threat from alcohol vapor might be much more. I personally think that the levels of such threats are benign. I believe I've shown the general validity of the comparison although I have not had my post of about 40 hours ago acknowledged by Mark C. yet.

What I believe is that alcohol at low levels is far less likely to be carcinogenic than SHS at low levels, because ethanol does not interact directly with DNA, and its carcinogenicity is most likely secondary to gross organ toxicity, in which case there is probably a threshold below which no carcinogenic effects occur. SHS smoke, on the other hand, includes components that interact directly with DNA on a single molecule basis, and therefore likely has no such threshold.

But what I believe is not the same as what I know. To know whether or not alcohol vapor is indeed benign, we would need real experimental or epidemiological data. It seems that your argument can be summarized as follows

1. Alcohol vapor is at least as carcinogenic as SHS (theoretically doubtful and unproved).
2. Alcohol vapor in bars is unregulated. (true)
3. Therefore alcohol vapor is relatively harmless (irrational and unproved)
4. Therefore SHS also should not be regulated. (irrational)

Posted by: trrll | July 23, 2007 11:18 PM

I don't know and I'm only asking a question -- so please don't stone me.... but you're making me curious -- isn't the suggestion that alcohol is related to mucosal cancers, which I've seen around (If I'm stating that incorrectly - feel free to correct) a suggestion that alcohol is more than only related to gross organ toxicity?

Mucosal surfaces can be considered an organ. But I probably should have said "cellular toxicity," since that is the basis of organ toxicity.

Posted by: trrll | July 24, 2007 7:55 AM

GDF:

The policies that could be based from applying the right models can be discussed ad infinitum, but those policies are outside of the models as such - they need their own models, if possible. (Type of society, health cost, and a lot more.)

You saw a jump in my argument when I referred to one of several possible policies in the given context as I needed to elaborate on the general background. But there are others possibilities and other contexts, as I pointed out.

trrll:

SHS smoke, on the other hand, includes components that interact directly with DNA on a single molecule basis, and therefore likely has no such threshold.

An experienced biochemist once pointed out to me DNA repair mechanisms as probably giving an apparent threshold for such mutagens - below their saturation the repairs are largely (though not fully) successful. Speculation, although informed, still doesn't substitute for epidemiological data though.

Posted by: Torbjörn Larsson, OM | July 24, 2007 9:54 AM

I agree Mr. Larsson. I did think you were jumping from this discussion to a political (policy) discussion and my questions were meant to get a sense of where you were coming from.

I aqree with trrll who, I think, earlier suggested that the *policy discussion* should be grounded in epidemiology, as opposed to bio-chemistry (although as I pointed out -- that's a hornet's nest as well).

trrll -- Thanks. I was specifically referring my question (about mucosal cancers) to Mark C's "Further, alchohol isn't a direct carcinogen - it's carcinogenicity is a secondary effect: large quantities of alchohol damage the liver, which can eventually lead to cancer." I understand that even if alcohol is implicated in, for example, nasal and throat cancer it *could* still be an indirect effect of ingestion -- but seems like it would suggest the possibility of something more direct. Your own statement didn't seem to be as *strong* about organ toxicity as Mark C's. (Again, bear with me please, this is not my field at all).

I have suggested before (based on Mark C's critique) that Mr. McFadden overlooked carcinogenicity as a variable in his example. But the more I think about it, the variable list seems almost endless. Humidity, different evaporation rates, behavior of gas vs. particulate, applicability of rodent carcinogenicity to humans... (Mr. Larsson just brought up teratogenicity) Days from now, I expect someone is going to post "yeah, well what if the bar was on the moon?"

I don't think the problem is that Mr. McFadden didn't go far enough in his analysis... I think he went too far, started down a complex bio-chemical road that perhaps wasn't necessary to his overall point (about safe levels of Class A carcinogens, as I understand it).

Posted by: GDF | July 24, 2007 11:29 AM

Torbjörn wrote:

"I don't agree with this being hysteria, since we are discussing epidemiological risks for bystanders. We could also add economical losses and losses in individuals freedom of choice which comes from creating lower quality environments, but that is another debate IMO."

But all of that is radically variable vis a vis SHS due to the range of SHS situations as was well expressed by GDF's #56 post above. The range of SHS situations is so great, in fact, that a zero tolerance, total smoking ban in all public places (and the technically true but highly misleading when unqualified assertion that SHS is "worse" than first hand smoke) does indeed amount to hysteria.

Torbjörn, if we should have zero tolerance for polution re SHS (I don't know if you're advocating zero tolerance toward SHS, but it sure *seems* that you are), why not for any # of other environmental pollutants, as well? Let's be consistent about pollution.

Posted by: Norm Breyfogle | July 24, 2007 12:57 PM

trrll -- Thanks. I was specifically referring my question (about mucosal cancers) to Mark C's "Further, alchohol isn't a direct carcinogen - it's carcinogenicity is a secondary effect: large quantities of alchohol damage the liver, which can eventually lead to cancer." I understand that even if alcohol is implicated in, for example, nasal and throat cancer it *could* still be an indirect effect of ingestion -- but seems like it would suggest the possibility of something more direct. Your own statement didn't seem to be as *strong* about organ toxicity as Mark C's.

I don't want to be dogmatic about it. I'm not an expert on alcohol metabolism, so while ethanol is unlikely to interact with DNA, I'm not absolutely certain that there is no point in alcohol metabolism that could lead to generation of free radicals that could react with DNA (and how about people with acetaldehyde dehydrogenase deficiency?)

Posted by: trrll | July 24, 2007 2:19 PM

I apologize trrll - I don't even know what acetaldehyde dehydrogenase deficiency is -- it may be completely relevant -- you understand I was just reacting to the form of the question?

Posted by: GDF | July 24, 2007 4:18 PM

Interesting. Mark C. has accused me not once, but at least a half dozen times on this single page (not to mention the denialism and insolence blogs) of being dishonest, employing a "strategy of deception" and of using "bogus arguments. Yet he has not shown *any* evidence to back that up.

He pointed out an oversight in my original analysis: my failure to take into account the comparative carcinogenicities of various carcinogens. In the process he himself pulled up examples of several carcinogens which are NOT known to be human carcinogens.

I responded with a full admission of the oversight and then, despite his own failure to do so, I provided my own analysis of the comparative carcinogenicity of smoke's Class A components and illustrated that the fundamental argument of my essential point was still quite valid.

Mark C holds a doctorate in computer science, and yet despite an obvious interest in discrediting me has yet to try even the simple experiment outlined in my original thesis. Since however that thesis was being questioned again, I repeated the experiment and my results do indeed support my basic contention: given normal humidity, temperature, and airflow levels the rate of evaporation of ethyl alcohol from a standard martini are going to be on the order of a gram as opposed to the less than 1/1000th of a gram of Class A components emitted by a cigarette. The only better "evidence" that I could offer on this would be cut and pasting the SGR and IARC pages themselves and providing a 48 hour long YouTube video of an evaporating martini.

(Note for experimenters btw... the original was done with grain alcohol to avoid the confounding effect of the differential vaporization qualities of alcohol and water. If you use gin or vodka instead you'll find that the evaporation rate is high at first and then slacks off as the alcohol concentration in the mixture is reduced. At the end of 48 hours the remaining liquid is probably going to be about 98% water although you would need to do a proper chemical analysis to actually prove this.) (See: http://en.wikipedia.org/wiki/Ethanol_%28data_page%29#Boiling_points_of_aqueous_solutions and http://en.wikipedia.org/wiki/Distillation and http://en.wikipedia.org/wiki/Azeotrope if you want to be "nitpicky" (LOL!) about it all.)

I would think that before someone with a professional degree would call another researcher dishonest they would at least attempt to check the work that they are criticizing.

He went on to say, " No one credibly believes that alcohol inhalation at a bar is dangerous. There is absolutely no evidence for that, whatsoever." and I quite agree with him. The same was true for secondary smoke until quite recently. Today, there *is* at least some evidence that being exposed to secondary smoke in normal quantities is "dangerous" in a loose definition of that word. The evidence itself is very poor scientifically but it does exist.

Mark goes on to say "No one credibly believes that alcohol inhalation at a bar is dangerous." I agree Mark, and, if it weren't for the elaborate campaign that has been mounted over wisps of secondary smoke for the past 30 years no one would "credibly" believe that secondary smoke inhalation at a bar is "dangerous".

I believe I have made my "stupid comparison" quite well, despite Mark's feelings on the subject. It is indeed a fairly unimportant one, but unlike the run-of-the-mill Antismokers out there I do try to back up what I say when challenged... even if the challenge seems designed more to be a time waster than a productive contribution.

As has been noted, quite correctly, this level of quibbling over every miniscule detail is "fundamentally dishonest"... but ah ent da one who brunged in da quibbles! :> The major point, that we are blithely overlooking the probably nonexistent threat of normal concentrations alcohol vapors in our demonic concentration on the also probably nonexistent threat of normal concentrations of secondary smoke stands firm.

Smoking bans are bad laws based upon lies and in my opinion a law based upon lies is no law at all... but that's a math of a different morality.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 24, 2007 6:01 PM

Torbjörn wrote:

"As I said to GDF just previous, a policy discussion is a much larger issue and not relevant to the question of McFadden's bad math. Apparently you think that a common regulation amounts to hysteria, while my guess is that it is done because epidemological studies has shown that SHS causes harm.

What common regulation? How strict? These details are of the essence of the question of "hysteria" re SHS regulations. If the common policy is a zero-tolerance one (i.e., no SHS allowed in any public places under any circumstances whatsoever, I can see how that could be logically classified as hysteria. (At the very least, it's inconsistent with the amount of other pollutants we readily allow into our environment.)

Throughout my comments on this thread I've maintained that McFadden's alcohol analogy is specious at best, but it does imply the wider issue of other environmental pollutants and the consistency in policy (or lack thereof) to which I'm referring.

Posted by: Norm Breyfogle | July 25, 2007 3:15 AM

But I agree with all that, Torbjörn. I'm merely pointing out that any such common regulation should be consistent for all environmental pollutants.

Posted by: Norm Breyfogle | July 25, 2007 12:53 PM

I'm merely pointing out that any such common regulation should be consistent for all environmental pollutants.

And I'm merely pointing out that any such common regulation also should be consistent for all other societal issues concerned, such as smoking being a recreational and habituating drug and who pays associated health costs. In effect, what trrll says.

So maybe there is no inconsistency though we won't agree on the contexts.

Posted by: Torbjörn Larsson, OM | July 26, 2007 10:13 AM

One thing still sticks in my mind, and that is the implication I often hear that SHS is more dangerous as it is typically encountered than is first hand smoke. Except under very controlled, highly unlikely, and pretty much never to be encountered stiutations, this bit of often implied hyperbole fits my definiton of hysteria. I remain open to being convinced otherwise, but no one has yet even come close to achieving this for me.

Posted by: Norm Breyfogle | July 26, 2007 8:49 PM

Norm, If you want to see a similarly destructive-to-science movement in its early stages:

Ahhh... second-hand-obesity.

http://abcnews.go.com/Health/Diet/story?id=3413751&page=1

Study: Obese Friends Could Make You Fat
Obesity May Spread Through Social Networks, New Research Suggests

"Using new software, Harvard and UCSD researchers created diagrams that plotted obesity and relationships, mapping the past 30 years.

Obesity and Socializing
While mapping networks of neurons in the brain or HIV prevalence among different communities has been popular in the past, this is the first time that obesity has been put under the lens of social networks.

The researchers found that when a person becomes obese, the chances that a friend will become obese increases by 57 percent. Siblings of obese people have a 40 percent increased risk of obesity, and their spouses' risk increased by 37 percent."

Posted by: GDF | July 26, 2007 9:34 PM

"could" --- or not. Just like alcohol. Which, as we've both said trrll, brings us back to the epidemiology.

Posted by: GDF | July 26, 2007 11:59 PM

trrll:

Excellent explanation. I've been searching for the right way to phrase exactly what you just said, and failing.

But that's exactly the point - you can't just extrapolate from the properties of what a smoker inhales; second hand smoke isn't identical to first-hand, and in some ways it's more dangerous than first hand. That doesn't mean that someone who is breathing second-hand smoke is in more danger than the smoker, because the SHSer is breathing *less* of it - but the *makeup* of the SHS is different. So a straight-line extrapolation of the effect of smoking to SHS doesn't necessarily generate an accurate assessment of risk, because it's *different*.

Posted by: Mark Chu-Carroll | July 27, 2007 8:54 AM

Coin:

Thanks, this is the type of factoid that stuck in my mind. Presumably an health agency must be able to back up their claims, but it isn't certain.

trrll:

I second Mark's acclaim.

Posted by: Torbjörn Larsson, OM | July 27, 2007 10:13 AM

I wood say that I unnerstud all the stuf abuv but I cant cuz I caut sum stoopid frum my frends.

Agh! What's next? Second hand rap stylings? Second hand religious belief? Second hand infidelity? Second hand patriotism? Second hand war fervor? Second hand zero tolerance for all correlated second hand effects?

Mark CC, second hand obesity is indeed just as silly as it sounds.

Posted by: Norm Breyfogle | July 27, 2007 1:48 PM

On the other side, you tend to find Jews and Asians performing significantly better in schools than their white Protestant classmates. That's not because Jews or Asians are any smarter than the WASP kids - it's because there's a strong social/cultural influence that pushes them to perform academically.

Are you sure? This sounds like one of those reassuring "just-so" stories that people like to tell themselves, because it is more comforting to believe that--in contrast to just about any other biological parameter that you can measure--intellectual capacity has no correlation whatsoever with one's genetic background.

Posted by: trrll | July 27, 2007 3:04 PM

That last sentence didn't make any sense, but I think you know what I meant :-)

Posted by: JBL | July 27, 2007 3:39 PM

Just realized the comment I addressed to MarkC was for Jonathan.

*sitting back while trrll and MarkC and JBL address the meaning of the correlation of ethnicity and academic achievement, and what social policies should stem from such a finding.*

Posted by: GDF | July 27, 2007 4:46 PM

saying that someone has a "goyisher kop" in Yiddish means literally that they have a "gentile head" -- gentile being the word for non-Jewish. What it actually *means* is that the person is stupid. That's one example a pretty damn massive sociocultural influence to dismiss as a "just-so story."

The question of whether there is a sociocultural influence is distinct from the question of to what extent it is causative. The two could be hard to separate, because it is likely that a culture would tend to promote those activities where they tend to inherently have an advantage due to genetic background.

Posted by: Anonymous | July 27, 2007 6:33 PM

Unfortunately for GDF, I don't really find the question of biological versus social cause very interesting: even if there is a measurable biological/genetic effect on some quantifiable trait which we can nickname "intelligence," it seems to me extremely unlikely that anyone will be able to develop a method to actually separate out the social and biological influences.

Actually, I expect that will happen fairly soon. We are rapidly approaching a point when individual gene sequencing will be practical on a large scale. It will then be fairly straightforward to correlate performance on various tests of cognitive performance with gene polymorphisms. It is hard to imagine that some alleles associated with cognitive advantages will not turn out to be more common in people of particular ethnic backgrounds, considering this has been the rule for other genetic polymorphisms that have been studied to date.

I am concerned that the dogmatic insistence that all of the variance in cognitive performance of people of different ethnic backgrounds must be cultural rather than genetic will leave us unprepared to deal with the political fallout from the almost inevitable discovery that this is not true.

Posted by: trrll | July 27, 2007 8:20 PM

Jonathan Vos Post and Mark CC,

I didn't write that the studies on social pressures toward obesity are crazy, just silly and stupid in SHS terms (which you agreed with, Mark CC). I don't doubt the social influence phenomenon at all, but in the context of this thread it is in fact a silly interjection. But at least it's hilariously funny to have it appear here.

Btw, anyone who's socially influenced to become obese or bigoted or superstitions or a war drum beater or whatever is in fact arguably displaying "stupidity." Remember, 50% of people are below average in any area. So the studies re social influence are actually studies of stupidity!lol(I know, I know, not stupid studies, but studies about stupidity. I'm just having fun here.)

The social influence phenomenon leads to all the stupidities I alluded to in my #97 post, and many, many more.

As for trrll's very ominous comments about genetics and intelligence, I must sadly agree. Seems we're all marching inevitably into a powerfully fascist future no matter what we do.

Posted by: Norm Breyfogle | July 28, 2007 2:26 AM

Someone on a related thread wrote about "the tactics" of public health. I posted the second-hand-alcohol press release(which fell into my lap like a gift from the gods at the right moment) to illustrate that point. The new "tactic" is to reframe individual issues as social issues. And, since individuals don't seem to be moving in the preferred direction, to bring the weight (no pun intended) of society to bear on the individual. Of course, we're people, so there is *some* social aspect to every individual act. So what now? Give up autonomy for the larger good of society? (How much harm is *too much*?) Should I insist that those around me lose weight? Do I have a *right* to that? For those with a sense of history -the reframing of individual choices as social responsibilities is frightening. (Well, that's a political discussion, I suppose).

But the larger problem, to me, is that public health isn't even "science" in the sense that you are discussing. It is not neutral. The analytical tools and methods of public health make results far too easy to manipulate, and the reality is that public health has become a cesspool of agenda (ideological, political, financial). A related issue - just a few years ago, I was cringing at the phrase "social marketing" and now it's quite common to see that phrase in public health research. We aren't informing or educating anymore. We're *marketing*. And apparently, we've adopted the methods of the used-car salesman.

I still believe, that Orac (I'm assuming he's a bright guy) took the wrong side of this SHS issue and failed to apply his critical thinking skills. That's all I'm asking anyone -- take a good, critical look at this *science* of public health (this goes beyond SHS) before leaping to its defense.

Good point Mr. Larsson - about "too much freedom" leading to repression. Seems anti-intuitive and took me a moment to *get* -- but I agree.

trrll -- I tend to agree that genetic influences are more important than we have *wanted* to acknowledge. What do you think that fallout would be?

The thing I like about THIS thread, is that at least we are discussing these issues, and not calling each other names. That's refreshing!

Posted by: GDF | July 28, 2007 1:14 PM

"Nor should we expect it to be, since politics of managing economics and morals around it unavoidable comes in outside scientific investigations as such."

Agreement there - if you were saying that the question of what to *do* with scientific results - is a political/moral question.

"(And sometimes threatens to come into them as well. :-( )So it isn't an added problem but "a challenge"."

I was suggesting that it's not longer a "challenge" - we're losing that battle, and it's a big problem. Like most in my field, I was used to being teased by my friends in the "hard" sciences -- suggesting that what I do is not "real" science. And like Orac on another thread, I argued that it was "real" science -- but that science-in-the-world was just harder to do. I've recently given up that argument. It's not that it can't be "real" science. It's that, I now believe, in practice it too often isn't.

If I was really on a rant, (*grin*) I would suggest that SAS and SPSS for Windows (as well as other modeling software) was the worst thing that could have happened. That's when (it seems) "researchers" really started generating numbers with no understanding for what they mean. No understanding of critical assumptions. No understanding of limitations. No thought to research design. It's a common joke that statisticians are brought in to mask the flaws in research design. It's not a funny joke anymore. Okay... so I'm a dinosaur. Or, as my colleagues started calling me -- a "purist". Wow -"purist" - what an insult.

*rant over*

"Not surprisingly these often small and mismanaged investigations turns out to be next to worthless, and harmful by adding "noise" to the picture or being taken as basis for suggesting erroneous actions"

Yes, see Glantz's "Helena" study.

trrll -- do I understand correctly that the same alcohol that we allow to evaporate in bars - we don't allow to evaporate in your lab? (Serious question -- I'm not sure what you were saying)

Posted by: GDF | July 28, 2007 2:52 PM

trrll wrote:
"Actually, I expect that will happen fairly soon. We are rapidly approaching a point when individual gene sequencing will be practical on a large scale. It will then be fairly straightforward to correlate performance on various tests of cognitive performance with gene polymorphisms. It is hard to imagine that some alleles associated with cognitive advantages will not turn out to be more common in people of particular ethnic backgrounds, considering this has been the rule for other genetic polymorphisms that have been studied to date."

We already have lots and lots of data correlating race to scores on certain tests -- it would be rather surprising if that correlation didn't also carry over to some genes which have varying presence over racial groups. I'm not sure why you think such a result would shock people, nor why you think there will be any "political fallout" from it. It also has little to do, as far as I can see, with the intent of my comment. In particular, I don't see how such a study could control for social influences any more than is currently possible, largely because social influences are extremely difficult to quantify. (This is aside from the fact that intelligence itself is difficult to quantify, of course.)

In any event, the obvious policies to deal with such potential problems, at least domestically, are those which promote integration and, indirectly, miscegenation.

Posted by: JBL | July 28, 2007 5:31 PM

trrll -- do I understand correctly that the same alcohol that we allow to evaporate in bars - we don't allow to evaporate in your lab? (Serious question -- I'm not sure what you were saying)

Correct.

Posted by: trrll | July 28, 2007 5:43 PM

Can we simplify this issue and just talk about gender differences? Seems like only dealing with (basically) two cases, and not having to deal with either mixtures (or integration) would be easier.

But trrll -- when such differences (intellectual, for example) were simply assumed (without biological evidence) - it did create some basis for discrimination -- No? Why would it be different with biological evidence?

And btw -- I completely agree -- I meant by my previous comment about not wanting to face this issue that the scientific community seemed to not want to -- I'm all for facing it square on.

Posted by: GDF | July 28, 2007 7:40 PM

Ahhh... I was making the assumption that one could never control for cultural difference anyway -- (even same sex siblings develop in different "cultures")

Thus, I thought the 2 case problem was simpler.

I also meant something more serious than to say that the within group variation is greater than the between group (which is what we say now when confronted with such differences).

I meant this -- suppose we find that there is, for example, a "math" gene (you know what I mean) that is distributed unevenly. That FAR more males have this gene than females. Certainly cultural influences such as education can affect the resultant mathematical ability -- but where educational resources (and genetic testing resources) are scarce -- should they go to the group that can make more use of them?

This is the type of question I think would occur.

Posted by: GDF | July 28, 2007 9:26 PM

the within group variation is greater than the between group (which is what we say now when confronted with such differences).

Yes. But it was my understanding that a PCA would still separate many characteristics between human populations, as allele frequencies differ due to migrations. Is that wrong?

GDF:

My perspective on public health, or evidence-based medicine (EBM), is a layman. So I am obviously ill equipped to understand the problems practitioners perceive, "purist" or not. :-P

But I'm encouraged by EBM, which seems to be a rather late invention. The methods are as "hard" as in other sciences, but as I understand it it is unfortunately not generally possible to get 3 sigma differences between modeled observations as we would like to see in physics. (Or 5 sigma signals for unmodeled de novo observations.) Often experimental physicists have easily adjustable "knobs" to amplify differences and disentangle causality.

That, and the often many confounding factors, must make biology and medicine difficult, akin to parts of astronomy or geology. (AFAIK just about the only non-overlapping distribution found is the difference in finger pressure sensitivity between newborn male and female humans.)

Posted by: Torbjörn Larsson, OM | July 28, 2007 11:02 PM

trrll -- okay - gender was NOT a good example. Hormonal confounding. You are correct.

You suggested that we might need to face at some time that there are significant genetic differences between groups of humans. And that by not facing it - we risk giving the fact undue importance. I suggest that we already know that there are cultural, and/or genetic differences that appear in statistical comparisons of populations (as you noted, for lack of evidence, we tend to explain these away as cultural)-- and that we then explain even cultural difference away (from serious political consideration) by invoking the greater variance within group mantra.

But -- I thought you were going for a more important question. That is, what about differences that are really differences? Population A has the math gene in abundance --not so for population B. We work out all the possible confounders -- and there sits the "math" gene (just an example - I don't care what kind of gene...) . What happens then?

So again, my question -- what is the fallout if we find a real (non, or hardly overlapping) difference between populations A and B? (And I think where Mr. Larsson was going was division into smaller and smaller sub-populations would make that likely) Are you saying that something like that could never be found? That what I called "the mantra" will always serve us?

Mr Larsson -- LOL! If I had it to do over again I'd be an astronomer -- just came in from watching a satellite go by... wonderful planet we live on.

Posted by: GDF | July 29, 2007 12:22 AM

I think where Mr. Larsson was going was division into smaller and smaller sub-populations would make that likely

Probably true, indirectly. Aside from the graininess that should result from allowing more dimensions of characteristics, there should be graininess by sufficiently specifying subpopulations according to migrations/allele groups as in the Genographic project. The purpose would be to maximize resolution when studying genomic effects, I think.

Of course, if we go all the way to sequencing all individuals we will see some differences (from genetic defects and what not), if not "populations". So the question becomes how we interpret any resulting subpopulations (size) and differences (strength).

The Genographic project would indicate that we should be at a finer level than the old "race" concept (for population size, if not strength), so the stigmata should be less. I hope.

[The Genographic project seems to take the tack of researching extended ancestry and its migrations. That is one way of making the subject more neutral and personal, allaying some of the concerns trrll raised perhaps. Evolution to the rescue!]

Posted by: Torbjörn Larsson, OM | July 29, 2007 9:20 AM

Where I'm going with this question -- or the broader context to me is this -- Most of us, I think, on some level believe that science supports those things we think are "good", and disproves those things we think are bad. Racism, for example. Or, in a political form, unequal distribution of resources by race - which seems somewhat immoral to me. But I'm asking -- Could science and morality ever conflict? Or is morality derived from science? Or could science lead us to make an immoral choice? (This isn't a trivial point about bad science - I'm talking about good science). If it's scientific can it also be immoral? Would we sometimes want (or have to) ignore science for the benefit of humanity?

Those are different approaches to the same or related questions.

That's the context -- but the question (in layman's terms) was much simpler. Could significantly unequal (little overlap) gene pools legitimize unequal distribution of resources?

Posted by: GDF | July 29, 2007 11:33 AM

There IS more variation within group vs. between groups. Hence the concept of "race" has gone the way of phlogiston, caloric, and the humours.

Yes. But I believe I have seen the issue of distinguishable smaller groupings in more detailed analysis (as in PCA) raised by some. Unfortunately, as of this moment, I'm unable to find it. So I should drop that part and concentrate exclusively on the nature vs nurture part.

GDF:

I'm just asking for an opinion. Yours too Mr. Larsson.

Sorry, I thought this was about the nature vs nurture question. Backtracking, it seems to be initiated by you, introducing the recent study on socialization of obesity. (Comment #89.)

Hmm. Not having studied the report, the result would certainly be consistent with what we would expect, if it can be ascertained. (One epidemic study with low power, as I assume it is, isn't going to convince me.) Confounding factors can be genetic (siblings) and diseases (people speculate in infectious agents as causes), though I would assume they correct for the former.

But I suspect your immediate question is contained in the new comment #131:

But I'm asking -- Could science and morality ever conflict? ... Would we sometimes want (or have to) ignore science for the benefit of humanity?

All questions about morality. I don't want to be the person who diverts this bad math thread into a large discussion on morals et cetera of the specific question here or on your followup on societal choice ("legitimize unequal distribution"), but these are certainly reasonable questions on science and its limits.

I don't see why science and morals can not conflict, since science is absolute (on observations) while morals are relative. On the other hand, science is part of the human project too, so it relies on certain morals. (Easily notable in medical research, for example.) You could as well ask if morals in one group can conflict with morals in another group. So it seems. :-P

Could the project of science have goals conflicting with its own morals? Yes, you wouldn't want to satisfy your curiosity by different types of harm, to patients et cetera, even if it could shorten the time to a remedy that saves more future lives. Difficult but vital questions, and, I hasten to add, questions that we all should consider and act on whether we are biologists or physicists or laymen.

The last question is different and more difficult in principle, it isn't exclusively a moral question. The moral questions hang together and concern science as it is practiced, but your two following questions jumps toward some mysterious goal. Also, I'm not sure I understand in what situations the last question would apply, even considering that it could lead into your followup question. Do you have an example?

Posted by: Torbjörn Larsson, OM | July 29, 2007 2:12 PM

note: I understand that Mr. Jefferson was making a moral statement, But the grounding for the moral argument conflicts (at least once, if not twice) with scientific observation.

Posted by: GDF | July 29, 2007 2:58 PM

I understand that Mr. Jefferson was making a moral statement, But the grounding for the moral argument conflicts (at least once, if not twice) with scientific observation.

More than in Jefferson's time? I think that Jefferson understood, for example, that some people are bigger than others.

Posted by: trrll | July 29, 2007 4:29 PM

Thomas Jefferson was a scientist, or close enough. A gathering of scientists in the JFK White House was dubbed the greatest collection of scientific minds there, since Jefferson dined alone.

As a scientist, Jefferson could NOT have meant: "We hold these truths to be self-evident, that all men are created equal" to be an assertion of mathematical equality of parameters of phenotype.

He presumably meant some combination of Jefferson's Three Laws of Constitutionobotics:
(1) All men are created equal UNDER THE LAW, as the USA would be a nation of laws, as described later in the Constitution, not of kings annointed by God;
and, hedging his bets cleverly as a Deist:
(2) all men are created equal IN THE EYES OF GOD, but which transcendental equality, being beyond human comprehension, would not be allowed to conflict with Law (1) above.
(3) To the extent that it does not violate Laws (1) and (2), no action or inaction shall allow the CONSTITUTION to be destroyed.

Am I close, Drs. Asimov and Chu-Carroll?

Posted by: Jonathan Vos Post | July 29, 2007 7:20 PM

Torbjorn wrote, "[Any parallels with McFadden's unreplicated experiment is coincidental, I'm sure. :-P]"

Torbjorn, is alcohol very expensive or illegal where you live? I would imagine it must be or you would surely have attempted replication of the experiment you find so questionable at some point in the last week or two.

You were proposing I perform an alcohol concentration test that would probably involve some quite expensive apparatus. I propose a trade: you provide that apparatus and upon receipt I will send you a 50g bottle of vodka for your replication efforts.

Meanwhile, I note that it's been two full days at this point since I answered Kevin Vicklund's statement that, "It has been 5 days since McFadden announced the start of his experiment. I have yet to see him post any results. Michael, please post the results of your experiment. Or did you not actually perform the experiment?"

I posted those results... but have not heard an acknowledgement.

Nor have I heard back from Mark C. after I performed the comparative carcinogenicity analysis he requested, using the sources that he provided while criticizing my "ridiculous mathematical error".

The lack of responses is eerily similar to the silence of the Helena authors when it comes to defending their own work in the pages of the British Medical Journal.

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 29, 2007 9:40 PM

GDF:

Um, okay - I'm not sure why you introduced 2nd hand obesity then, and I don't think I should go into the generals of moral statements and politics here.

Mark and trrll was discussing a similar social situation, and the problem with untested ad hocs. Which unfortunately is still a large part part of "nurture vs nature" debate. (That is something I feel comfortable to discuss without going completely OT, btw.)

But I could discuss political mistakes on science:

But the grounding for the moral argument conflicts (at least once, if not twice) with scientific observation.

Not in the complete context which you shopped off - it is about giving populations at large the same general rights. What you are discussing is if differences, whether cultural or biological would imply differences on top of that.

Well, obviously. Objective differences would be support or exceptions for groups that needs that. (For an example that comes to mind, you can't use car safety belts or collision bags on too young people, they need their own protection. So if you have law mandating belts of bags, you need exceptions.)

But I don't see that it applies to those general rights as described in your quote - so from that perspective all individuals are equal. (There is no "fib" - what you are discussing is another context IMHO.

[At the worst you can accuse the declaration of being incomplete. But that seems unfair, because it doesn't concern these contexts. To give a complete and definitive description of even empirical phenomena and models is impossible, at least in everyday texts, even less subjective areas. And I believe one of the goals with the declarations were to be as clear and brief as possible.]

Or in other words, pretty much what trrll and Jonathan Vos Post has already noted.)

Posted by: Torbjörn Larsson, OM | July 30, 2007 3:00 PM

Torbjorn wrote, "Btw, going over the comments someone (Stephen) did a similar experiment and got an evaporation rate of an alcohol/water mixture (heavy on the alcohol) of ~ 0.1 g/h (comment #39). That differs from yours of ~ 0.7 g/h by the famous factor of 10."

Indeed it does, which was why I invited others to replicate it as well. In Stephen's case we were dealing with a situation of relatively high humidity and, as he noted, the liquid was at the bottom of a beaker and only 40% alcohol with probably greatly reduced surface air flow compared to what would be seen in the original experiment involving a martini glass with grain alcohol in a room with moderate air flow. I'm sure Mark C and Orac and others would have found the proposition of enough interest to at least give it a whirl, yet we've heard nothing from them... which would indicate to me a reasonably strong possibility that they tried it, found the results to be unsupportive of their stance, and yet were honest enough not to deliberately misrepresent their findings.

T, you go on to say, "So it seems you don't understand the phenomena well yet. And that is the task you have taken on yourself."

I believe I understand the phenomenon quite well: alcohol is a highly volatile Class A carcinogen. The only discussion at hand is just how rapidly you are giving your fellow diners cancer. I believe you are not. The "no safe level of exposure" folks would argue otherwise.

As for "Gish galloping" (the tactic of reeling off dozens of unsound arguments knowing that time constraints preclude refuting them all)," I wasn't aware I was reeling off "dozens" of unsound arguments here.

Please name a dozen or so... or even six. Or how about just naming and refuting three... I'm sure the "time constraints" would allow for that.

Michael J. McFadden
Author of "Dissecting Antismokers' Brains"
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | July 30, 2007 6:49 PM

Nice to see Michael finally post his results (I had to go out of town over the weekend). Sad to say, the experiment was rather poorly done. If I taught chemistry, his efforts would have netted him a Fail and an invitation to conference hours to discuss falsification of results.

Not having access to exact measurement equipment (my multibillion femtodollar grant from PM only bought me a used teaspoon from the junky on my front porch) I would have to be honest and say that I started both glasses with roughly 50g of liquid. After about 50 hours I found about 35g left in the pure water glass and slightly less than 15g left in the 50/50 glass.

Roughly? About? Slightly less? How did you come by these measurements? What scale did you use to determine the masses, and what were the gradations and errors of the scale? Or did you just eyeball it and say "that's about one-third of what I started with"?

But let us continue under the assumption that your data is accurate.

If someone here would like to stand up as more of an expert in chemsitry and dispute me they are welcome to, but based upon my previous experimentation with 95% ethanol, I believe that about 15g of water evaporated from the water glass glasses along with about 20 to 25g of ethanol from the 50/50 glass along with 10 to 15g of water. While I did not take time-measured interval measurements to determine rate of evaporation, I believe that the 20-25g of alcohol largely evaporated during the first 24 hours of the experiment... thereby validating my "roughly 1 gram per hour" statement.

Sorry, but your claims show both a lack of knowledge about chemistry and an impermissible assumption in your conclusion. As you were warned before, you can not use a positive azeotrope such as 95% grain alcohol to determine the rate of evaporation of a different concentration of the mixture. To be more accurate, the evaporation rate of a positive azeotrope is the maximum possible rate for those two substances, whether pure or mixed with each other. The further the concentration is from the azeotrope, the slower the evaporation rate of each substance.

Therefore, if the water glass lost 15g to evaporation, the mixture had to have lost more than 15g to evaporation. Therefore, less than 20g of ethanol was lost to evaporation. Furthermore, it is unacceptable to simply assume that all the alcohol will evaporate in the first 24 hours, when that is the very measurement you are trying to ascertain!

We can't conclude anything from the data as given other than that after 48 hours, less than 20g of ethanol has evaporated. That gives us a rate of about 400 mg/hr with huge error bars. There's not a whole lot that can be concluded from so shoddy an experiment.

One thing that has been mentioned is the problem of measuring alcohol content. An inexpensive method is to use a hydrometer. It won't be as accurate as a mass-spectrometer, but it probably would give a usable approximation.

Oh, another thing I should point out (which has already been mentioned). Surface area is an important component of the evaporation rate. The smaller the surface area, the slower the rate. So using glasses with curved or angled sides will give a time-variable rate. Using a big glass like a tumbler will give a faster rate than using a small glass like a shotglass.

Posted by: W. Kevin Vicklund | July 30, 2007 7:07 PM

I cannot be objective about this, due to the number of people in my family who died of cancer, including my mother, my father, my maternal grandmother, my father's sister, and many more. None of those which I listed smoked cigarettes, except that my father briefly did, lighting them with a Zippo filled with fuel from the airplane he used to teach piloting to Free French pilots in World War II.

My father's sister and her husband both died of bladder cancer, and both enjoyed a daily martini or two.

I wonder about the effect of industrial carcinogens on my mother and her mother, from all those years in Elizabeth New Jersey.

As to Bad Math in biology, there's a some nice analysis in:

arXiv:q-bio/0701039 (replaced) [pdf, other]
Title: Universally Sloppy Parameter Sensitivities in Systems Biology
Authors: Ryan N. Gutenkunst, Joshua J. Waterfall, Fergal P. Casey, Kevin S. Brown, Christopher R. Myers, James P. Sethna
Comments: Submitted to PLoS Computational Biology. Supplementary Information available in "Other Formats" bundle. Discussion slightly revised to add historical context
Subjects: Quantitative Methods (q-bio.QM); Molecular Networks (q-bio.MN)

Posted by: Jonathan Vos Post | July 30, 2007 9:02 PM

McFadden:

Since so many have elucidated how and why your experiment is unrepeatable, shoddy and ultimately yours, I don't need to continue.

[But I note that Kevin Vicklund, who seems to have a fair grasp on chemistry as acute as on plate tectonics and gravitational calculations, also criticizes mixed evaporation. (Which I, probably incorrectly, referred to as "assisted". Kevin seems to discuss from deviation from an azeotropic state instead. Yup, I should not dismiss vapor pressures.)]

Please name a dozen or so... or even six. Or how about just naming and refuting three...

How about that being effortlessly galloping a dozen or so... or even six... or finally three... nitpicks, which are completely besides the subject at hand?

And before you diagnose that as, fairly, a cheap rhetorical trick (albeit provoked by the context of Gish galloping being a cheap rhetorical trick) if it is without support - yes, what about the points Mark pointed out?

Posted by: Torbjörn Larsson, OM | July 31, 2007 1:22 PM

Even if it's true that there is "no safe level of exposure", (and of course it must be true to some iota of degree, at least), this same reasoning should apply to every other pollutant in our environment, e.g., petroleum emissions, radiation of various sorts, excess fat and other contaminants in our food,and yes, even alcohol evaporation and the common cold (not to mention second hand obesity and other social conditioning ills). Saying that this fairness principle I'm advocating would only hold in a "perfect world" dodges the rational issue involved and doesn't suit a rational blogsite like this one.

No, this is not correct. "No safe level" applies when a single molecule of toxin is capable of producing lasting damage by acting all by itself. This is pretty much exclusive to molecules that chemically react directly with DNA to produce mutations. Most toxins do not work this way--their mechanism of toxicity is indirect, and requires cooperative interactions between a large number of molecules acting simultaneously in different places, producing an extreme nonlinearity in the concentration-effect relationship that is referred to a "threshold." When the concentration is well below the threshold, there is essentially no risk, and this can therefore be referred to as a "save level."

Posted by: trrll | August 1, 2007 10:18 AM

So, one molecule of an SHS carcinogen is "unsafe," but the issue of consistent public policy toward other carcinogens is irrelevant?

I rest my case.

Posted by: Norm Breyfogle | August 1, 2007 12:54 PM

Norm:

I don't see how this issue can be discussed without immediate implications re social policy. Hence, the science is clouded by political agendas. The debates on this thread alone amply illustrate that fact.

I think that is a completely unfair characterization, since a few here, among them you, are trying to take the discussion outside the factual one on the science into the subjective and contingent one on social issues. So you are among the one in that illustration, and the rest is a neat counter-illustration.

one molecule of an SHS carcinogen is "unsafe,"

How did you get that out of a long discussion of different kinds of exposures and the conclusion: "So there *are* things for which there is no safe level of exposure. Is cigarette smoke one of those? Honestly, probably not."?

Intriguing. :-P

Posted by: Torbjörn Larsson, OM | August 1, 2007 2:53 PM

Just so ya know I ain't gone 'n runned away on y'all... :) Bit busy with other things at the moment, but WILL respond soon to the comments made. Kevin made some very good points: quite possibly strong enough to completely counter my contentions about the alcohol experiment and statements: I'll have to look at them more closely to respond.

Mark, you continue to slather your responses with "idiotic", "stupidity", "bullshit" and such, but at least you've backed off a bit from accusations of dishonesty. I appreciate that: if you examine all the thousands of my posts over the years on webpages/blogs/newsgroups I don't think you'll find a single one where I was dishonest or would even leave a reader with a strong impression that I was dishonest. I also don't think you'll find any where I simply ran away from a superior opposition argument.

Those arguing against smoking bans, with the rare exception of events like last November when the tobacco companies got involved with promoting idiotic "weak bans", have virtually or literally no money for the battle. Our strength HAS to lie in our honesty... it's our best weapon against the lies and exaggerations of the other side. Am I claiming that NO ONE on my side of the issue is extremist or dishonest? Of course not. But I can speak for myself and I do.

OK! That's all I have time for at the moment... but in my best Arnie voice, "I'll Be Back!"

:>
Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | August 1, 2007 6:51 PM

MarkC - The issue Norm points to of why SHS seems to be singled out (why A and not B) is not a case of "simple" inconsistency in regulation. When you see A and not B I suggest that you should ask yourself why. Is there a powerful pro B lobby? Is there a powerful anti A lobby? And if so, what is bought-science doing to public health?

To borrow your phrase, I believe that the campaign against SHS (no matter how unpleasant you personally find SHS) is mostly "facile nonsense present with a huge amount of spin, misdirection, and babble."

It may be the first of such campaigns -- but if scientists don't look at it critically, and start to demand more of public health it won't be the last.

It's a big issue -- not the kind you can just post a link on a blog and the matter is settled (I can't anyway). So all someone like me can say is, if you really care about the scientific integrity (as opposed to you're just happy to have the smoke gone) please take another look at what you are claiming or defending.

If you are really interested, I suggest that a good place to start would be to look at both sides of the issue around the research of James Enstrom, rather than just one side. (I believe Orac already presented the ACS view) - here is Enstrom's defense. It's interesting reading about the workings of public health even if you aren't that interested in the issue of SHS.

http://www.scientificintegrityinstitute.org/defense.html

Enstrom's is a very long statement -- but I hope readers here have long attention spans - because it's a very important statement.

Posted by: GDF | August 2, 2007 12:11 PM

Mr Larsson -- I've tried to provide the evidence for my concern here and on the blogs related to this one.I've posted links to the statements of other scientists, like Richard Carmona, and Richard Smith --who are very concerned about the direction public health has taken. I've tried to give explanations and examples to support my statements. I've also tried to take seriously the comments of others, and engage in fruitful discussion. I've also (many times) tried to make the point that this is about more than SHS.

That's all I can do.

I'm unclear about whether you meant your last sentence to refer to me -- or to Dr. Glantz's attack on Dr. Enstrom. (as per the link I provided)

Posted by: GDF | August 2, 2007 11:46 PM

Repeatability means consistency in science, not just that you can repeat it.

A more understandable description is perhaps that since we are concerned with the result, repeatability obviously means that we can repeat (replicate) the measurement (observation, data), not that we can repeat the measurement procedure.

Posted by: Torbjörn Larsson, OM | August 4, 2007 1:17 AM

Aside from continuing his accusations of dishonesty several more times Mark brought up the point of why I spent so much time "defending" my experiment. The defense was precisely because I do not feel it is proper to simply ignore criticisms of defects in one's work the way the antismoking folks do: I stand behind what I say, will defend it where it needs defense, and openly admit any defects it has when they are properly pointed out... as I believe I have done here.

Mark goes on to make several points that I feel are faulty. To avoid accusations of "Gish Galloping" I'll stick to simply addressing all three bulleted ones he made:

"- You have shown no, zero, zip, nada evidence to support any claim that inhaled alcohol is a carcinogen."

Actually, I showed the IARC references, and then, after correction by Kevin, admitted that they indicate that it is possible that OTHER elements than alcohol may be responsible for the Class A designation of alcoholic beverages. In terms of comparing the effect of inhaled vs. sipped alcohol on human mucousal membranes, I believe the effects would be far more scientifically defensible as similar than the data Mark wanted to use comparing data from all sorts of tissues of rats and mice to human beings.

"- You continue to insist on an "experiment" using evaporation of grain alchohol, even though grain alchohol doesn't evaporate at the same rate as alchohol in solution with water."

I insisted upon the honesty of my initial experiment, and repeated a near variation of that experiment when the results were questioned. I have become aware that the evaporation rates are somewhat different, but given the overall quantity comparison of the initial experiment (2,000 times, or 200,000%) I feel that in this particular aspect of the argument that Mark seems so concerned about a difference of 20, or 30, or even 50% is not particularly of great concern. I offered to rectify this difficulty upon provision of the proper equipment to do so. I do not feel the "nitpicky" (as I believe Mark characterized it) point is important enough to warrant the investment on my own.

"- You continue to discount by pure hand-waving the idea that second-hand smoke might be dangerous. You *admit* that it contains carcinogens; the entire argument that it's not a risk is by faulty metaphor with alchohol."

Not the entire argument or pure handwaving at all Mark, and while I would certainly hesitate to return your own comments about honesty, I must admit I find it difficult to believe that you've kept yourself from any possible learning during all this discussion by visiting:

http://www.nycclash.com/Philly.html#ETSTable

or the websites linked through my own signature for more background. Still, I'll grant the possibility that maybe you're just not into reading stuff and weren't aware of the broader arguments.

As for the question about being "paid by Big Tobacco"... it would have been kind of silly to lead off my book the way I did if I was. You can betcher bottom dollar I would have been "exposed" by now after being such a pain in the butt to the antismoking lobby. LOL!

Heh... I didn't even join Forces or the Smokers Club all through the 90s because of my own suspicions that they might have BT connections and I didn't want my own work/efforts to be "tainted" by that association. Nowadays, after realizing the amount of money being pumped into the other side and seeing the relative dearth of funding to fight it I'm not so sure if our grass-roots groups really *should* be so concerned about their purity. We could certainly accomplish a helluva lot with about 1/10th of one percent of what the Antis get. The problem of course is that BT money has to be "discounted": a dollar from BT doesn't produce a dollar's worth of results precisely because no one (including about 98% of us smokers' rights activists) trusts them further than they could throw them. If BT came out tomorrow and said the sky was blue, most of the country would probably decide it must be pink and purple polka dotted instead.

So in brief, no Jonathan, I think you'll have to opt for either insane or rational. :>

Michael J. McFadden
Author of "Dissecting Antismokers' Brains"
http://pasan.TheTruthIsALie.com

Posted by: Michael J. McFadden | August 4, 2007 10:14 PM

Hmm.... sent this out about 30 hours ago but it doesn't seem to be up here yet so I'll try again:

======

Aside from continuing his accusations of dishonesty several more times Mark brought up the point of why I spent so much time "defending" my experiment. The defense was precisely because I do not feel it is proper to simply ignore criticisms of defects in one's work the way the antismoking folks do: I stand behind what I say, will defend it where it needs defense, and openly admit any defects it has when they are properly pointed out... as I believe I have done here.

Mark goes on to make several points that I feel are faulty. To avoid accusations of "Gish Galloping" I'll stick to simply addressing all three bulleted ones he made:

"- You have shown no, zero, zip, nada evidence to support any claim that inhaled alcohol is a carcinogen."

Actually, I showed the IARC references, and then, after correction by Kevin, admitted that they indicate that it is possible that OTHER elements than alcohol may be responsible for the Class A designation of alcoholic beverages. In terms of comparing the effect of inhaled vs. sipped alcohol on human mucousal membranes, I believe the effects would be far more scientifically defensible as similar than the data Mark wanted to use comparing data from all sorts of tissues of rats and mice to human beings.

"- You continue to insist on an "experiment" using evaporation of grain alchohol, even though grain alchohol doesn't evaporate at the same rate as alchohol in solution with water."

I insisted upon the honesty of my initial experiment, and repeated a near variation of that experiment when the results were questioned. I have become aware that the evaporation rates are somewhat different, but given the overall quantity comparison of the initial experiment (2,000 times, or 200,000%) I feel that in this particular aspect of the argument that Mark seems so concerned about a difference of 20, or 30, or even 50% is not particularly of great concern. I offered to rectify this difficulty upon provision of the proper equipment to do so. I do not feel the "nitpicky" (as I believe Mark characterized it) point is important enough to warrant the investment on my own.

"- You continue to discount by pure hand-waving the idea that second-hand smoke might be dangerous. You *admit* that it contains carcinogens; the entire argument that it's not a risk is by faulty metaphor with alchohol."

Not the entire argument or pure handwaving at all Mark, and while I would certainly hesitate to return your own comments about honesty, I must admit I find it difficult to believe that you've kept yourself from any possible learning during all this discussion by visiting:

http://www.nycclash.com/Philly.html#ETSTable

or the websites linked through my own signature for more background and where you'll see a great variety of arguments completely unrelated to this fairly minor point. Still, I'll grant the possibility that maybe you're just not into reading stuff and weren't aware of the broader arguments.

As for Johnathan's question about being "paid by Big Tobacco"... it would have been kind of silly to lead off my book the way I did if I was. You can betcher bottom dollar I would have been "exposed" by now after being such a pain in the butt to the antismoking lobby. LOL!

Heh... I didn't even join Forces or the Smokers Club all through the 90s because of my own suspicions that they might have BT connections and I didn't want my own work/efforts to be "tainted" by that association. Nowadays, after realizing the amount of money being pumped into the other side and seeing the relative dearth of funding to fight it I'm not so sure if our grass-roots groups really *should* be so concerned about their purity. We could certainly accomplish a helluva lot with about 1/10th of one percent of what the Antis get. The problem of course is that BT money has to be "discounted": a dollar from BT doesn't produce a dollar's worth of results precisely because no one (including about 98% of us smokers' rights activists) trusts them further than they could throw them. If BT came out tomorrow and said the sky was blue, most of the country would probably decide it must be pink and purple polka dotted instead.

So in brief, no Jonathan, I think you'll have to opt for either insane or rational. :>

Michael J. McFadden
Author of Dissecting Antismokers' Brains
http://pasan.TheTruthIsALie.com

Posted by: MIchael J. McFadden | August 5, 2007 10:25 PM

I agree with the GDF's post as well.

I'm not a doctor, a biologist, or a statistician. I'm not even a scientist in any real or strict sense, but it seems clear to me that biological organisms exhibit extremely wide ranges of resistance to carcinogens and other pollutants, making it very difficult to achieve rational consistency in public policy regarding their regulation, given the various political/industrial realities in modern societies. Genetics, personal diet and exercise, possible psychological variables, all together appear to affect the statistical risks as much as does any universally applied public bans of SHS specifically (of course, even this statement is arguably true or false to some degree).

As I wrote far above, there's a pretty large grey area here in which reasonable people can reasonably disagree.

If one doesn't smoke, eats well, and gets good amounts of exercise, I suspect that most SHS effects are pretty insignificant, or at least as insignificant as are the effects of many other modern industrial pollutants, even for those who work regularly in typical SHS environments.

Of course, I could be wrong.

Posted by: Norm Breyfogle | August 6, 2007 4:05 AM

GDF:

Bans are not without social and economic costs of their own, but alternative public policy solutions are never considered.

This is a two way street - bans usually incur gains, which is why they are considered. For example, banning smoking totally would benefit health and economy. But it isn't considered due to historical reasons.

So while you lament some bans, always remember that other people laments the lack of others.

we aren't talking mustard gas here.

Not relatively speaking, but this is again a two way street. Mustard gas was banned because people suffered intolerable. Many more have suffered intolerably from cancers by smoking.

It is again the scope of the context that matters. SHS is an involuntary secondary effect to a voluntary habituating drug. [Do you understand why I think this would be a lengthy discussion? :-P]

I would argue that we discuss the facts of the case instead of the scope of the context and the social solutions. Alas, the main fact here is that McFadden is a kook. ;-)

one breath of smoky air and I'm a goner

That wasn't the claim, which a short reread of the thread would show you. Or perhaps you don't get that the measured statistics applies to populations, not individuals.

Either way this just emphasis my impression that this discussion have long since left its purpose of debunking bad SHS claims.

Posted by: Torbjörn Larsson, OM | August 6, 2007 8:52 PM