And even if the experimenter were able to offer such vast sums of money to his subjects
      that every last one of them agreed to comply with the required drinking regimen - and no
      experimenter has such resources - then two things would happen: (1) the subjects would
      cheat, as by many in the zero-glass group sneaking drinks whenever they could, and many
      in the many-glass groups drinking less than was required of them; and (2) subjects who
      found their drinking regimens uncomfortable would quit the experiment. Subjects
      quitting the experiment constitutes a fatal blow to experimental validity because it
      transforms groups that started out randomly constituted (and thus equivalent in every
      conceivable respect) into groups that are naturally constituted (and which must be
      assumed to be probably different in many conceivable respects) - a conclusion that I
      will not pause to explain in detail.
      Manipulating long-term alcohol consumption in an experiment is unethical. And we know
      that no such experiment has ever been conducted because it would be unethical to conduct
      it, and would inevitably lead to the experimenter being sued. That is, it is unethical
      in scientific research to transform people's lives in possibly harmful ways. Most
      specifically, it is unethical to transform people's lives by inducing them to drink
      substantial amounts of alcohol every day for several decades. The potential harm is
      readily evident.
      For example, drinking 10 glasses of wine per day, or even several glasses, will
      predispose a person to accidents. A single experimental subject who consumed several
      glasses of wine and then was incapacitated in an automobile accident would be all that
      it would take to bring such research to a halt forever. The accident victim might
      readily argue that the experiment requiring him to drink wine was responsible for his
      accident, and that the experimenter - and the university at which he worked, and the
      granting agency that funded his research - were liable for millions of dollars. In
      anticipation of no more than the possibility of such a law suit, no granting agency
      would fund such research, and no university or research institution would allow it to be
      conducted under its roof.
      Consuming substantial amounts of alcohol can not only cause accidents, but it can also
      ruin health, destroy careers, distort personalities, break up marriages - for which
      reason no experiment will ever require subjects to consume substantial amounts of
      alcohol over extended periods of time. The possibility of harm, and thus of law suits,
      can even be conceived at the low end of the alcohol-consumption continuum. That is, a
      subject prohibited from drinking any alcohol might argue that this for him unnatural and
      unaccustomed regimen changed his personality, undermined his career, and ruined his
      marriage, and with this claim in hand, could readily find a lawyer willing to help him
      sue for damages.
      And if such an experiment had ever been conducted, it would
      be invalid
      Manipulating long-term alcohol consumption in an experiment would fail to meet the
      double-blind requirement. And although we are certain that an experiment manipulating
      alcohol consumption over an extended period has never been conducted, even if it were
      conducted, it would nevertheless contain inescapable flaws which would stand in the way
      of permitting cause-effect conclusions. For example, you may be aware that the best
      experiments are ones that are "double-blind." A "blind" experiment is one in which the
      subjects do not know what experimental condition they are in - they might not know, for
      example, whether the pill they are swallowing contains a curative drug, or only a
      placebo. In our alcohol experiment, they would not know whether the liquid they were
      drinking was wine, or only some wine-colored and wine-flavored water that had been
      sealed in wine bottles. Already, we see the impossibility of our wine experiment being
      even so much as blind. Just about every subject in our wine experiment would
      immediately realize what it was that he was drinking. Tinted water is clearly
      distinguishable by its appearance and taste and effect from wine. A blind wine
      experiment, then, is an utter impossibility. Most subjects would be able to quickly
      infer approximately what experimental condition they had been placed into.
      A "double-blind" experiment would be one in which neither the subject nor the
      experimenter knew what experimental condition any particular subject was in. For
      example, the experimenter hands the subject a capsule, but does not himself know until
      the experiment is over whether that capsule contains a curative drug or only a placebo.
      In our alcohol experiment, a double-blind experiment would involve the experimenter
      monitoring the life and health of each subject, but only after the experiment was over
      opening up the sealed envelope to find out how much alcohol that subject had been
      consuming over the past 30 years. Utterly impossible as well.
      The reason that the double-blind requirement is essential is that without it,
      confounding factors appear that might be responsible for any observed longevity
      effects. For example, subjects aware that they are in a large-alcohol-consumption group
      would also tend to realize that such alcohol consumption might harm them, and so they
      might attempt to compensate by taking vitamin pills, not smoking, upgrading their diets,
      exercising, and so on. Or, they might start eating fats prior to drinking alcohol, in
      order to coat their stomachs and slow the absorption of the alcohol. They might do a
      large number of things. What is important is that the knowledge of one's experimental
      treatment can lead to one or more changes in behavior, and that it is these unintended
      changes, and not the wine consumption itself, that could affect longevity, either in one
      direction or the other.
      Or, here is a particularly plausible confounding that might appear. Imagine that the
      experiment attempts to control wine drinking, and no more than that, and that subjects
      do faithfully follow the wine regimen that is imposed on them. Nevertheless, the less
      wine that they were allowed to drink, the more beer and hard alcohol they would probably
      end up drinking, but which would make the initially equal groups unequal on beer and
      hard-alcohol consumption. And so then it would be impossible to tell if differences in
      longevity should be attributed to differences in wine consumption, or to differences in
      beer consumption, or to differences in hard-alcohol consumption.
      But while we may choose to pause and speculate as to what confounding variables may
      appear, scientific method does not obligate us to do so. We know that confounding
      variables are possible in non-double-blind experiments, and the number that we are able
      to imagine is limited only by the time that we allocate to trying. If I cared to spend
      a few hours thinking about it, I could write several pages of possibilities. If I chose
      to spend a few months thinking about it, I could write a book of possibilities. I am
      able to imagine confounding variables either improving health or impairing it at the low
      end of the alcohol-consumption continuum, and as well either improving or impairing
      health at the high end of the alcohol-consumption continuum. Scientific method does not
      require us to know for certain what and how many confounding variables may appear to
      destroy the validity of an experiment which is not double-blind; rather, scientific
      method assures us that it is so likely that one or more confounding variables will make
      their appearance in a non-double-blind experiment, that such an experiment must be
      considered to be fatally defective, and that no cause-effect conclusion can ever be
      drawn from it with confidence.
      Thus, no valid experiment exists. In short, we can be sure that no experiment has ever
      been conducted to ascertain the effect of long-term alcohol consumption on longevity,
      and that if such an experiment had ever been conducted, the impossibility of its being
      double-blind, or even blind, would render it inconclusive.
      The French Paradox Research
      Must Have Been Correlational
      But if the data featured in your 60 Minutes broadcast was not experimental, then what
      was it? It must, by default, have been correlational. That is, rather than subjects
      being assigned randomly to groups and being required to drink a given volume of alcohol
      each day, it must have been merely observed what volume of alcohol they chose to drink
      each day.
      Alcohol consumption would be measured by self-report. Well, it is not quite true that
      the experimenter would observe what volume of alcohol his subjects drank daily. It
      would be impractical to follow subjects around and actually see how much alcohol they
      consumed in restaurants, in bars, in their homes. Much more likely is that every once
      in a long while, the subjects would be mailed a questionnaire asking them to report how
      much alcohol they had been drinking lately. The inability to measure alcohol
      consumption directly is already a weakness - subjects might not remember accurately how
      much they had been drinking, or they might experience some pressure to distort how much
      they had been drinking either upward or downward. However, this is not at all the big
      weakness that I want to bring out, so let us get to that without further delay.
      We have already seen that random assignment guarantees pre-treatment equality on all
      dimensions. I first recapitulate that in the case of the random assignment of subjects
      to groups in an experiment, we were guaranteed that the subjects in each group would be
      initially equivalent on every conceivable dimension. The larger the random groups, the
      closer to being precisely equal on every conceivable dimension would they become. Thus,
      in a properly designed and executed double-blind experiment, any differences that
      subsequently arose between groups would have to be attributed to the different
      treatments that the experiment had administered to them - for example, if some groups
      lived longer than others, nothing else would be able to explain this except that some
      groups had consumed a different volume of wine than others.
      Natural assignment guarantees pre-treatment inequality on many dimensions. But in a
      correlational study, subjects are not assigned to groups randomly, they assign
      themselves to groups naturally. A subject who is in a no-wine group, for example, is
      one who has himself decided that he does not drink wine. Thus, the groups are referred
      to not as randomly constituted, but as naturally constituted, as if nature had come
      along and assigned each subject to one of the groups. Now here comes the really
      important part. It is that experience teaches us that naturally-constituted groups are
      capable of differing from each other on every conceivable dimension, and are highly
      likely to differ from each other substantially on a number of dimensions. In other
      words, people who drink no wine are likely to differ from people who drink several
      glasses of wine in many ways. Perhaps the non-drinkers will have more females, and the
      drinkers will have more males - or perhaps the opposite. Perhaps the drinkers will be
      older or younger. Perhaps the drinkers will be richer or poorer. Perhaps the drinkers
      will tend to be single and the teetotallers tend to be married, or vice versa.
      Differences may readily be discovered in height, in weight, in education. Differences
      could quite plausibly be discovered in smoking, in drug use, in exposure to industrial
      pollutants, in diet. People who drink will tend to live in different parts of the city
      from people who don't drink. People who drink may watch more television, use microwave
      ovens more, spend more time breathing automobile exhaust - or less. As people of
      different ethnic backgrounds, or religions, or races drink different amounts, it follows
      that people who drink different amounts will differ in ethnic background, in religion,
      and in race.
      One can speculate about thousands of ways in which drinkers could differ from
      teetotallers, and if one actually examined two such groups, one would find a few
      dimensions on which such extraneous differences were large, several dimensions on which
      such extraneous differences were moderate, and a large number of dimensions on which
      such extraneous differences were present but small. The hurdle that the correlational
      researcher is never able to overleap is that given that he is unable to look for every
      conceivable difference, he will never know all the ways in which his
      naturally-constituted groups did indeed differ from each other.
      Natural groups may eat different amounts of broccoli. And so then, no cause-effect
      conclusion will ever be possible from a correlational study. If the moderate drinkers
      happen to live longer, we will never be able to conclude that this is caused by their
      moderate drinking, because it might be caused by how close they live to high-voltage
      lines or how often they wash their hands or how far they drive to work or how much
      toothpaste they swallow or how much they salt their food or how close they sit to their
      televisions or how many pets they keep or whether they sleep with their windows open or
      whether they finish their broccoli. In an experiment, random assignment of subjects to
      groups guarantees equality on all such extraneous dimensions, and this makes
      cause-effect conclusions possible. In a correlational study, natural assignment of
      subjects to groups guarantees inequality on many such extraneous dimensions, and this
      makes cause-effect conclusions impossible.
      Correlation does not imply causality. Every textbook on statistics or research
      methodology underlines this same caveat, captured in the expression "correlation does
      not imply causality," which warns that from correlational data, it is impossible to tell
      what caused what. Science has developed only a single method for determining what
      caused what - and that method is the experiment. No experiment, no cause effect
      conclusion - it's that simple. Given correlational data, furthermore, there is no way
      of extracting cause-effect conclusions by more subtle or more advanced analyses - no way
      of equating the groups statistically, no way of matching subjects to achieve
      statistically the pre-treatment equality that is needed to arrive at cause-effect
      conclusions. Advanced methods of analyzing correlational data do exist, and are used by
      naive researchers, and to the layman may appear to be effective, but the reality is that
      all are fatally flawed, all have been demonstrated in the literature to be ineffective
      and to lead to inconclusive results. The bottom line is that there is no way to extract
      cause-effect conclusions from correlational data.
      You overlooked that the causal direction might be reversed. In the case of The French
      Paradox finding, I can readily see a plausible alternative interpretation as to how the
      observed data could have arisen. The data do seem to show that as drinking declines
      from a high to a moderate level, longevity increases. This accords with the notion that
      alcohol is toxic, and that its effects are deleterious. What constitutes The French
      Paradox, however, is that when one goes even farther along the drinking continuum from
      moderate drinking all the way down to no drinking at all, instead of longevity
      increasing still higher, the opposite happens - longevity shrinks.
      What distinguishes the scientifically-trained mind from that of the layman in this case
      is that the layman thinks of a single interpretation, and seizing on that as the only
      one possible, stops thinking. That is, the layman thinks "Drinking not at all is
      unhealthy, therefore I can improve my health by drinking." The scientifically-trained
      mind, in contrast, recognizes that in correlational data a large number of
      interpretations is possible, acknowledges the first interpretation that springs to mind
      as one among the many that are possible, and keeps looking, and keeps finding, a number
      of alternative interpretations, and ultimately acknowledges the impossibility of
      choosing among them.
      As illustrated in my own case. Specifically, I happen to find myself in a
      naturally-constituted zero-alcohol group. That is, I drink not at all, or very close to
      not at all. There is a reason for this, and that is that the effects of alcohol upon me
      are toxic. Mainly, I get splitting headaches, even from the ingestion of small amounts
      of alcohol, particularly if the alcohol comes in the form of wine. I take this to mean
      that my constitution is weak, that I am unable to process alcohol efficiently, that I am
      unable to detoxify my body of alcohol the way that others can, that my body chemistry is
      not up to par. In other words, I am unwell, and as a result I do not drink.
      Please mark well what I have just done - I have reversed the cause-effect conclusion
      that you had come to. You concluded that not drinking causes deteriorated health, but
      what I am proposing to you at the moment is that deteriorated health can cause not
      drinking. The insight that I offer you is that when we observe a correlation, we don't
      know what caused what, and one of the possibilities to be considered is that the causal
      direction may be the opposite of our first impression, that a situation in which we
      first conjectured that A causes B may prove upon more thoughtful examination to be a
      situation in which B really causes A. In short, it may be the case that people who are
      destined not to live as long as others tend to find themselves unable to drink alcohol.
      That's all that the French Paradox may have discovered, and that's not a very good
      reason for anybody to follow your recommendation to go out and start drinking.
      Common sense alone invalidates The French Paradox conclusion. In other contexts, a
      correlation being misinterpreted to mean that drinking promotes either health or
      longevity will be obviously laughable. For example, a researcher who observes that
      hospitalized patients don't drink will not conclude that teetotalling causes
      hospitalization. Or, a researcher who visits death row and discovers that the inmates
      don't drink and do have short life expectancies will not conclude that teetotalling
      shortens life. In such examples, anyone with a modicum of common sense instantly
      recognizes that a correlation between zero wine intake and either poor health or short
      life does not mean that zero wine intake causes either poor health or short life. All
      that is required to recognize the invalidity of your conclusion in The French Paradox is
      to apply this same common sense to an only slightly more subtle case.
      Are there not other studies? Undoubtedly there exist in the literature a large number of
      studies that have some less direct bearing on the question that we are discussing, and
      many of these studies will be genuine experiments which do permit cause effect
      conclusions. I am thinking in particular of experiments that may demonstrate that
      ingredients found either in grapes or in wine have a certain physiological effect. With
      respect to such other studies, I make the following observations: (1) Your chief
      conclusion was based not on such experiments, but on one or more correlational studies.
      (2) An experiment in which subjects ingest an ingredient of grapes or of wine may
      witness a certain effect, even while actually eating grapes or drinking wine produce a
      different or an opposite effect. This could happen because in whole grapes or in real
      wine, the ingredient with the beneficial effect could be offset by some other ingredient
      which has a harmful effect, as by pesticides or nitrates that might be found in wine, or by the alcohol itself in wine. Unless an experiment actually has subjects drinking
      wine, no conclusions concerning drinking wine are possible. (3) An experiment
      demonstrating a physiological effect of something ingested is likely to be of short
      duration, and is not likely to measure the effect on longevity. However, demonstrating
      a physiological effect that appears to be beneficial (say a heightened level of HDL, as
      mentioned by Kim Marcus above) is not the same as demonstrating increased longevity,
      since the relation between the observed effect and longevity is speculative.
      In short, the only research that can prove that prolonged drinking of three to five
      glasses of wine per day can extend life is the non-feasible experiment that we have
      already discussed above in which subjects are required to drink different amounts of
      wine over an extended period of time, and the effects on longevity noted.
      The Harm That You May Have Done.
      What the above reasoning leads us to, then, is that you were without justification for
      promoting the conclusion that you did - that drinking three to five glasses of wine each
      day extends life. Quite possibly, your conclusion had the effect of increasing the
      consumption of alcoholic beverages, particularly wine, and possibly, the effects of this
      increased consumption have been uniformly bad.
      These may be among the damaging effects of your advice. The level of alcohol
      consumption that you advocate slows reaction times and interferes with coordination and
      impairs judgment, and therefore invites accidents. Certainly no airline pilot would be
      permitted to consume a fraction of your recommended daily intake and still be allowed to
      fly, and certainly every driver should recognize that he is putting himself at risk
      drinking as much as you advocate. We recognize the damage that your advice may have
      inflicted when we take into account that except for infants and the aging, accidents are
      the leading cause of death.
      The level of alcohol consumption that you advocate interferes with, or makes quite
      impossible, difficult mental work. Thus, a university student who follows your advice
      and has a couple of glasses of wine with his dinner is finished for the day - he might
      as well head out to a pub after that, because he will find his calculus homework quite
      incomprehensible. A chemistry professor who follows your advice and has a couple of
      glasses of wine with his lunch will find himself making mistakes as he tries to lay out
      the electron configuration of aluminum for his class - he had better find some simpler
      topic to treat in that lecture if he doesn't want to embarrass himself in front of his
      students. A lawyer arguing a complex case who follows your advice and has a couple of
      glasses of wine with his lunch will find himself losing the thread of his argument in
      court - he had better let his junior take over that afternoon if he wants to maintain
      his reputation.
      The level of alcohol consumption that you advocate may damage health. The level of
      alcohol consumption that you advocate possibly saps energy and depletes motivation,
      possibly leads to more time spent in small talk and in television viewing, and less in
      productive work and creative effort. Undoubtedly, the level of alcohol consumption that
      you advocate promotes outright alcoholism. Yours has been a call based on
      pseudo-science to abandon sobriety and embrace intoxication - hardly a direction that
      American culture needs to be pushed in.
      The French Paradox and The Ugly Face of Freedom were equally flawed. And to return to
      the comparison of your 23Oct94 broadcast The Ugly Face of Freedom to your 5Nov95
      broadcast The French Paradox, I do see a striking parallel. In both cases, you didn't
      know what you were talking about, but stepped forward and talked anyway. Given that you
      had not studied the subjects to which you addressed yourself, given that you had not
      thought about them, given that you were capable of nothing better than passing along the
      most superficial, man-in-the-street, off-the-top-of-my-head conclusions, the truly
      remarkable thing is that you would have the arrogance to think yourself worthy of
      standing up in front of tens of millions of people and telling them what was your
      opinion. Yet that is what you did, and in each case, you got it wrong. Your many
      conclusions in these two broadcasts ranged from totally opposite to the truth to totally
      unsupported by the evidence. The Ugly Face of Freedom for which you will always be
      remembered in the Ukrainian community was wrong and destructive. The French Paradox
      which judging from its Internet prominence appears to be your best-remembered broadcast
      among your total audience - was also wrong, and also destructive.
      A word concerning self-help. If you yourself subscribe to the prescription of drinking
      three to five glasses of wine each day, then I would recommend that you attempt to break
      yourself of the habit, and substitute for the many hours of inebriation thus avoided
      some sober study. Had you substituted for many hours of inebriation the sober reading
      of history, you might have spared yourself the fiasco of The Ugly Face of Freedom. Had
      you substituted for many hours of inebriation the sober study of scientific method, you
      might have spared yourself the fiasco of The French Paradox. Perhaps you have no more
      than to look at these two pratfalls in your own career to see how damaging is the effect
      of making a habit of indulging in alcohol.
      Disclosure would be a step toward restoring professional credibility. As enthusiasm for
      your French Paradox broadcasts seems to have its source in the wine industry, and as
      your integrity has been brought into question on the matter of The Ugly Face of Freedom,
      I wonder if your professional standing would not be enhanced by your assuring 60 Minutes
      viewers that you have received no benefits from the wine industry in gratitude for the
      increased sales that your French Paradox broadcasts have brought it. The absence of
      such an assurance will invite some 60 Minutes viewers to construe your French Paradox
      broadcasts more as infomercials than as investigative reporting.
      Lubomyr Prytulak
      cc: Ed Bradley, Jeffrey Fager, Don Hewitt, Steve Kroft, Andy Rooney, Lesley Stahl, Mike
      Wallace.
      HOME DISINFORMATION PEOPLE SAFER 1553 hits since 26Apr99
      Morley Safer Letter 8 26Apr99 One out of 40 escaped shooting
      It looks very much, Mr. Safer, as if on your 60 Minutes broadcast of 23Oct94, The Ugly
      Face of Freedom, your chief witness testifying to Ukrainian collaboration with the Nazis
      was himself a war criminal of substantial proportions, a former Gestapo agent with the
      blood of many on his hands, perhaps much of it Jewish blood.
      April 26, 1999
      Morley Safer
      60 Minutes, CBS Television
      51 W 52nd Street
      New York, NY
      USA 10019
      Morley Safer:
      I bring to your attention the following excerpt from an article by L. A. Ruvinsky
      published in the Ukrainian Historical Journal in 1985:
      After the end of the Second World War, the former head of the Lviv
      Gestapo, P. Krause, replying to a question put by the writer V. P.
      Bieliaev, testified: "If on our side, in the Gestapo, there had not
      worked several agents from among the Zionists, we would never have been
      able to capture and destroy such a large number of Jews, who were
      living under false documents and assumed names." For example, in July
      1941, Zionist Simon Wiesenthal, together with 39 other representatives
      of the Lviv intelligentsia, found himself in prison. Somehow, as a
      result of a "mysterious confluence of circumstances" all the arrested
      except for himself were shot, and he was freed. It is not surprising
      that after this, this Zionist provocateur became a regular Nazi agent.
      Polish journalists have established this as an indisputable fact. That
      is why the Hitlerites did not throw Wiesenthal into prison, which he
      frequently confirms, but rather sent him there to organize subsequent
      provocations. Evidently he was not lying when he said that he passed