Wikipedia talk:Requests for mediation/Passive smoking/archive 001

Mediator notice

I'm back up to speed. Again, I apologize for the delay & inconvenience, and I am thankful for your understanding & kindness. So, to get things rolling a bit, what's been going on with the disagreement over the past week? What is the current state of things? Vassyana 20:04, 13 August 2007 (UTC)

Chido6d (talk · contribs) is drafting a rewrite of the "Controversy" section at Talk:Passive smoking/Controversy draft, though he has asked the rest of us not to edit it for now. The biggest bone of contention is how much WP:WEIGHT to give the controversy and how to present the "minority" viewpoint. MastCell ^Talk 20:22, 13 August 2007 (UTC)

Hi Vassyana, welcome back. I faced the same situation as you just 4 days ago with the tragic and untimely death of my nephew. So, my heart goes out (now from experience).

Things have been quiet, but with a little reading you can catch up.

I did place the compromise draft on the special page. You can see comments there.

I would prefer that other editors draft their own than to criticize and edit mine. Constructive feedback is welcome, however, and has already led to a couple of changes. Chido6d 03:14, 14 August 2007 (UTC)

My condolences. Do you need a little time? Vassyana 07:00, 14 August 2007 (UTC)

Thanks very much. My time is limited for the next few days, but I'll still keep an eye on this and contribute. So, let's press on.Chido6d 10:34, 14 August 2007 (UTC)

Lets all do our own drafts first and then figure on the final version. And as for the minority viewpoint, we aren't arguing something like America didn't land on the moon, the challenge to the scientific consensus in this case has serious evidence and I believe your treatment of the matter is disparaging to fair discussion. Mickeyklein 05:10, 14 August 2007 (UTC)

Please focus on the content, not the contributors. Thanks! Vassyana 07:00, 14 August 2007 (UTC)

Proposal

It has been suggested each editor work on an individual draft. This could also work with a team of two or more contributors working on the same draft. Draft spaces could be created as sub-pages of this talk page. Just remember to use {{draft}} on your workspaces. After drafts are completed, we can compare, critique and work out an acceptable compromise/composite, within Wikipedia policy and guidelines. What do people think of this proposal? Vassyana 07:00, 14 August 2007 (UTC)

Thanks for this proposal. If you allow me, I'd like to make another preliminary proposal. It is to create a separate article in which we put all the "controversy" stuff, and to include a very short paragraph of a few lines in the main article saying that some scientists disagree with the consensus view and challenge the findings about the health effets of secondhand smoke, and that over time the tobacco industry has used various policies to undermine the results showing a link between diseases and secondhand smoking. Then we could even take the draft proposed by Chido6d as the first approximation of the new article, inserting a POV tag at the top, and then let the Wikipedia magic do its work, and in 6 months, the "controversy" page will have settled into a page acceptable to most.

For me, this is also the only way to preserve the WP:WEIGHT rule, since, the section on the so called "controversy" gives undue weight to unrepresentative views and very anecdotal facts, such as the reservations that Gori has with confidence intervals, and I fail to see how we will be able to shorten it, given the opposing views that we have as editors on these questions.

It might be good to agree on this proposal before each of us embark on proposing a draft, since this decision might change the way we write our drafts.

--Dessources 09:29, 14 August 2007 (UTC)

I accept the proposal of the mediator, and I welcome Mickeyklein to co-author a draft. I further welcome the advice and commentary of MastCell for representation of the opposing view.

I reject the second proposal for various reasons, including the suggestion that a POV tag be placed outright. It also results in what I believe is called a "fork". This fails to ultimately deal with the issues at hand.Chido6d 10:44, 14 August 2007 (UTC)

Creating a content fork has significant pros and cons, and shouldn't be undertaken lightly. On the one hand, it would be preferable to organize a neutral, properly weighted handling of the controversy on the main passive smoking page. On the other hand, the controversy is a slightly larger topic, with elements of public health, regulation, the tobacco industry and its tactics and litigation, etc. So perhaps a short summary in passive smoking and a somewhat more detailed treatment in, say, secondhand smoke debate or passive smoking controversy would be the way to go. I'll withold judgement on whether a fork is justified until I've seen what kind of proposed "Controversy" sections we're working with. MastCell ^Talk 17:48, 14 August 2007 (UTC)

I'll chip in for the draft of course. I should have some proposals in a day or two. 69.181.208.181 19:21, 15 August 2007 (UTC)

I'm bogged down a little with an ongoing ArbCom case, etc, but I am interested in putting a draft of my own together also. MastCell ^Talk 22:30, 15 August 2007 (UTC)

(Outdent) A few people have limited time, so I'd just ask everyone to be patient while draft proposals are put together. Thanks! Vassyana 05:27, 16 August 2007 (UTC)

I'm also busy, but I lean towards supporting a controversy page, which might also include some material from Smoking ban. This has worked reasonably well for global warming and other topics where the mainstream scientific viewpoint is subject to criticism for one reason or another.JQ 02:38, 18 August 2007 (UTC)

The objections Chido and I are technical in nature. They address the raw science of passive smoke and their effects based on experimentation following the guidelines of the scientific method. This material belongs on the page examining the scientifically verifiable effects of passive smoking.

Global Warming attracts a crowd of activists particularly opposed to the scientific method. I wouldn't be suprised if falsifiability was bullied out of existence on the main page. I will not be bullied like that. Mickeyklein 15:18, 19 August 2007 (UTC)

Please be careful to avoid original research. Any analysis of the raw science must be drawn from a published reliable source. We cannot make our own interpretations of data and findings, but we're instead required to report what available reliable sources state making sure that claims are represented in proper proportion. Also, please assume good faith on the part of your fellow editors. There doesn't appear to be any kind of "bullying" going on here. Be well! Vassyana 16:14, 19 August 2007 (UTC)

I do think Mickey's comment points up a bone of contention: there seems to be a desire to re-argue or re-debate the "raw science", epidemiology, and scientific methodology underlying the finding that passive smoking is harmful. This has occasionally run up against WP:SYN and WP:NOR, in my opinion, since the conclusions advocated by Mickey run counter to the interpretation established by a large volume of reliable secondary sources (e.g. Surgeon General, WHO, CDC, etc). MastCell ^Talk 17:15, 19 August 2007 (UTC)

What about as a general principle, one paragraph per serious objection (not including quotations). I dont know about Chido, but as far as space and proportion I would only want three, one for hypothesis testing objections (CI and dose response), one for publishing bias in meta-studies and one for the WHO report which I imagine would require its own section to properly explain the subtelties.

As for reliable secondary sources, I have them for all the objections. I have all the Lexis and JSTOR so I can search as we go along. I understand the need for verifiable sources and will try my hardest to provide them for all proposals. Mickeyklein 01:57, 20 August 2007 (UTC)

I'm sure you've already seen this, but it's worth looking at the distinction between primary sources (e.g. journal articles) and secondary sources (interpretive findings of groups of experts) set out in WP:NOR. I don't object to giving space to serious objections which are still current. I do object to giving a lot of space to objections which have since been discredited or are no longer represented among the views of experts on the subject, as if they were still current, particularly if we're just re-presenting primary-source journal articles in isolation with no secondary source to indicate how widely held the view is at present. MastCell ^Talk 02:22, 20 August 2007 (UTC)

I believe that statistical evidence that fails the p=.05 condidence interval is suspect, if that is a conspiracy theory idea I don't know what to say. Chido probably has something more eloquent in mind. 69.181.208.181 08:33, 20 August 2007 (UTC)

I know that you believe that, but the question for Wikipedia's purposes, per WP:WEIGHT and WP:NOR, is how widely accepted that view is among the scientific community. In U.S.A. v. Philip Morris, Philip Morris presented only one expert witness to dispute the scientific evidence on passive smoking. That witness, Edwin Bradley, took a line very similar to yours, arguing that any study with a p>0.05 was suspect. Here's the Court's decision:

Dr. Bradley's "methodology" depended on his evaluation of statistical significance. His position was that any epidemiological study whose result is not statistically significant must be discarded and cannot be relied upon to determine whether an association exists. Dr. Bradley stated:

"If a purported association is not statistically significant, your inquiry can end there."

No scientific or medical authority shares Dr. Bradley's view. Statistical significance is not one of the Surgeon General’s criteria for causality. As described below, it is a statistician's term of art, a tool to evaluate the possibility of chance in a particular study. Dr. Bradley's testimony confirms that even he recognizes this.

Moreover, Dr. Bradley admits that he stands alone in adopting and applying his test. When confronted with the conclusions of Sir Richard Peto, Sir Richard Doll, the Surgeon General, the EPA, WHO, IARC, National Research Council, and the American Heart Association, Dr. Bradley

responded...^{[1], p. 1235}

So it's pretty clear that, aside from one tobacco-industry expert witness (whom the Court deemed "not credible"), no scientific or medical authority shares Mickey's view. The question, then, is how much weight to give it? MastCell ^Talk 17:24, 20 August 2007 (UTC)

I thought I should chime in here, since I haven't been very active in this RfM lately. Anyway, I agree with MastCell, and I'll add that Mickey's statement that "I believe that statistical evidence that fails the p=.05 condidence interval is suspect, if that is a conspiracy theory idea I don't know what to say. Chido probably has something more eloquent in mind. 69.181.208.181 08:33, 20 August 2007 (UTC)" is the crux of the dispute. Chido and Mickey do not seem understand that Wikipedia is not to a forum to refight debates, present one's own analyses of data, or synthesize primary sources into novel conclusions. Until they can accept what wikipedia is and is not, I doubt that we'll get very far with this mediation. Sorry to be so negative. Also, I'm not a huge fan of a seperate controversy page as proposed by User:John_Quiggin.

I have begun work on a new draft, check it out here. So far I have mostly just reorganized things and have yet to attempt the (harder) task of trimming the fat to so as to bring the disputed sections in line with WP:WEIGHT.Yilloslime 21:50, 20 August 2007 (UTC)

MastCell, with all due respect, I caution you against original research. Dr. Bradley may have been the only one who testified at the trial; by no means does that imply that he is the only one who disagrees with the scientific majority. Actually, the most profound thing about Judge Kessler's opinion is her statement that statistical significance is not a criteria for causality...rather, it is "a term of art". I appreciate her use of the term art, because the kind of pictures they've been painting are most certainly borne of fantasy and deep creativity. After all, an excuse is surely needed when the numbers just don't add up.

That being said, I'm opposed to moving a section on controversy to its own page. The question at hand is this: Is it sufficient to include a disclaimer that a minority view is just that -- a minority view? Or is there a deliberate attempt to dumb it down or to silence it? Chido6d 02:28, 21 August 2007 (UTC)

I know, I know... everyone except you and Mickey is out to suppress The Truth. Look, Kessler's opinion was based on the testimony of numerous experts on biostatistics and epidemiology. Dismissing it with sarcasm doesn't really work. The Court's opinion repeatedly emphasized that "No scientific or medical authority shares Dr. Bradley's view." That's not WP:OR, that's a direct quote from a verifiable, reliable source. MastCell ^Talk 05:29, 21 August 2007 (UTC)

What about including more of Professor Peto's testimony? He speaks as Professor of Medical Statistics and Epidemiology at the University of Oxford, and apparently, he also believes that the evidence is not strong enough to make quantitative judgements. Mickeyklein 17:28, 21 August 2007 (UTC)

Fun quote of the day, from the old times of 1980 before the tobacco control movement:

"...a publication of the International Agency for Research on Cancer (IARC) which states "relative risks of less than 2.0 may readily reflect some unperceived bias or confounding factor".29 "

http://tobaccocontrol.bmj.com/cgi/content/full/15/suppl_4/iv68

The article was talking about the tactics of the tobacco industries expert tactics. Statistics is so $@#%^#$ unfair. Mickeyklein 17:39, 21 August 2007 (UTC)

You are really twisting in knots to avoid the fact that there's a scientific consensus. You're also mischaracterizing Peto's opinion. He believes unequivocally that there are health risks associated with passive smoking. However, like a small number of other epidemiologists, he feels that the magnitude of that risk is very difficult to accurately quantify. He is cited in Kessler's opinion as testifying to risks of passive smoking (see [2], p. 1235), and his testimony before the House of Lords, he stated:

I think there has got to be some risk... When you place the same mix of chemicals [as with active smoking], more or less, in to the general air, it must cause some risks, so I think the statement that there is some risk to non-smokers is well-founded; you would have to use the most extraordinary, implausible metaphysical arguments to argue for threshold doses below which there is zero risk... The definite statement is that some people are killed by breathing other people's smoke, and then there is reasonable uncertainty about the number killed.^[3]

Pretty clear. He believes passive smoking kills, but that exact quantification of the excess risk is very difficult. Elsewhere in his testimony, Peto points out that a 20% excess risk (as seen, approximately, in the WHO studies) makes sense from the standpoint of proportionality. You're trying to quote him to support an argument which he himself is not making. MastCell ^Talk 17:47, 21 August 2007 (UTC)

"Q401 Lord Skidelsky: Could I draw you out on one further thing. You have been unwilling to quantify the risks from passive smoking.

Professor Sir Richard Peto: Yes"

This does not match with the "overwhelming" and "infallible" evidence held by the scientific "consensus".

I believe a giant flying spaghetti monster causes lung cancer. But when I do the regressions I get insignificant results and I cannot quantify the risk involved. If only I had FDA approval then it would be true... Mickeyklein 18:14, 21 August 2007 (UTC)

Yes, that is a great example of taking a quote out of context. I provided the context above. Peto believes that passive smoking kills people, but feels that exact quantification of that risk is unsatisfactory. Even that is a minority position, though one that is well-supported enough to warrant mention in the article. "Overwhelming", as a descriptor of the consensus that passive smoking causes lung cancer and kills people, is quite accurate and Peto himself signs on, at least that far. As to your sarcasm, it's distinctly unhelpful. MastCell ^Talk 18:22, 21 August 2007 (UTC)

Problem is that the supposed consensus expounds reams of numbers that are supposed to show existing, quantifiable risks from ETS. If we were to take Peto at face value it would mean that the statistical claims of the consensus are unfounded.

And as for the sarcasm, I'll put it in more semantic terms. The absence of evidence cannot justify a truth claim. If the risks cannot be quantified then there is a lack of evidence and considering the base of evidence (absence) no claim can be made to the nature and origination of the risk. Mickeyklein 19:02, 21 August 2007 (UTC)

Not at all. Peto's argument is that the evidence undeniably supports the existence of harms from passive smoking, but that the effect size of those harms is uncertain. You can take that at face value or not, but for Wikipedia's purposes it is undeniable that consensus exists. As I said, there is a notable minority view that the size of the harms is unclear, but I have yet to see any evidence of a notable current view that passive smoking does not cause cancer etc. In fact, I've presented voluminous evidence that there is no notable scientific dissent on that point. MastCell ^Talk 19:12, 21 August 2007 (UTC)

A statistical truth claim cannot be unquantifiable. That is self contradictory. A truth claim that the lung cancer risk from ETS is greater than zero requires quantitative evidence that the risk is greater than zero. After all, the risk could be less that zero, as indicated in the IARC study for childhood exposure.

This implies that testimony such as Peto's disagrees with the scientific truth claims of the consensus. The consensus holds as a truth claim that the risk for ETS exposure in lung cancer is greater than zero and is surely not below zero. They back this up with statistical studies that claim a quantitative link.

Peto claims, on the other hand, that the risk is not quantifiable, and as it is not quantifiable the statistical evidence that shows the risk to be greater than zero and surely not below zero is unfounded and cannot support the consensus claim. Mickeyklein 19:40, 21 August 2007 (UTC)

The sources are quite clear. While you may disagree with them, we need to focus on what they actually say rather than putting our own gloss on things. MastCell ^Talk 21:36, 21 August 2007 (UTC)

A few requests

• Please refocus on the draft work. A demonstration of what you each would prefer to see will provide solid reference points for further discussion.
• Avoid debating the subject.
• Please ensure claims are verifiable in reliable references.
• Make sure to adhere to a balance according to available reliable references.
• Please double-check claims against reliable sources, so original research is avoided.

Thanks! Vassyana 07:45, 22 August 2007 (UTC)

I'd like to see a clear statement of the scientific consensus, at somewhat greater length than in the present version, and a brief description of the main criticisms made by the tobacco industry, its hired experts and consultants and its ideological (mainly free market/libertarian) allies. This can all be done with verifiable, reliable sources: the main requirement, as you say is to adhere strictly to WP:NOR.JQ 11:42, 22 August 2007 (UTC)

I should very much like to work on the draft, in the spirit proposed by Vassyana. However, I am facing pressure to finish some professional work commitment and have to postpone the drafting work by one week or two. In the mean time, I will follow the debate, and I may intervene, but without making a substantial contribution.

--Dessources 15:10, 22 August 2007 (UTC)

What may be failing to take hold here is the spirit of compromise. This is why I wrote my draft; I want to declare here and now that there is a lot about it that I do not like. But I scaled things back and included quite a bit of debunk material for the sake of those who disagree with me. Can we all at least agree that we are going to need to give in to some degree?

At the bottom line: why must some material go unchallenged, and any opposing view be attacked from all sides within the article text? Chido6d 01:20, 23 August 2007 (UTC)