Blog powered by Typepad

« An ordinary hero | Main | Why I'm not mad about Mad Men »


no one – like everyone else, I talk about “scientific facts.” Perhaps I was too glib, but it depends on how precise you want the terminology to be.

You are right, though: all science is based on probability, it’s just that some things are more probable than others; in fact there are some things (gravity, say) that are so consistently reliable that it would be perverse to lose sleep over the possibility that it might fail and we would find ourselves floating into space tomorrow.

Lightning strikes may, indeed, have another component, but psi is unlikely. A documentary I saw recently featured an Oklahoma farm worker who had been struck six times in the last thirty years. But because he works outdoors in all weathers, he is, by the very nature of his job, exposed to a greater risk of lightning strikes than the rest of us, particularly because Oklahoma suffers about a million lightning strikes per year. There are a number of factors that can skew the probabilities.

Michael – I am wondering what you mean by “statistically impossible” If a probability can be calculated, then you have a figure that states a probability, however remote it might be. If I tell you that an event has a probability of one in ten to the power of sixty eight, would you tell me that that event will never happen? (That is a mind-bogglingly large number, by the way.)

In fact, if you deal a full pack of well-shuffled playing cards (all of them, one by one), there are almost 10^68 possible combinations that could be dealt. Is anyone prepared to say that any combination dealt to them could not have happened? My point, again, is that if a psychic can predict such a hand in advance, then I will accept that precognition is real.

It is pointless to try to do any number crunching with regard to the person who won the lottery five times; there is not enough information available. The farm worker I mentioned earlier was increasing his chances of being hit just by virtue of the fact that his job puts him at higher risk anyway, in a state that has probably the highest rate of lightning strikes. If someone (and it happens) spends every spare money they have buying lottery tickets, then that person has a better chance of winning than someone who buys one ticket once or twice a year just on a whim. But suppose a lottery winner can now afford to buy hundreds of tickets at a time, which increases his or her chances of a further win. And if it happens, then that person might start buying tickets by the thousands or even tens of thousands. We just don’t know what has happened in those individual cases. It isn’t too difficult to work out “static” probabilities, but all variables have to be accounted for before one can have the confidence to say that a certain outcome is more likely to be due to psychic intervention than just pure dumb luck.

I think that if the lottery operators have any qualms about someone winning several times, they are more likely to suspect a glitch in the system or even the possibility of outright fraud before they go for the psychic hypothesis.

sbu – As you say, “The intention with my calculation was indeed to show that winning four jackpots is unlikely to happen by chance.”

I would put it another way: a genuine psychic who predicts the winning numbers in a lottery is unlikely to do so by chance. But I maintain that when anyone wins the lottery (even multiple times) it is the fact of prediction BEFORE the event that would make the difference, and convince me that there is something psychic going on.

Roger Knights – I alluded to your comment earlier, and I agree: without the full facts and information, which is unlikely to be discovered, the exact probabilities cannot be calculated.

Kris – I respect Paul and The Major. They are two commenters who have had something sensible to contribute to this ongoing debate about the paranormal, even though they and I disagree – sometimes strongly. Perhaps, like you, they dislike me for daring to disagree with them, but they have had the good manners not to make personal attacks against me, and nor have I made personal attacks against them, or anyone else. If you think that insults are a valid substitute for rational argument then you make yourself into a caricature of the woo proponents so beloved of those skeptics who are rather less charitable than me. And will you stop blathering on about “the law of the excluded middle?” It is a term that tends to be used in formal academic texts; if you mean a false dichotomy, just say it. There are educated people who regard those who use big words they don’t understand as intellectually stunted poseurs. And you wouldn’t want that.


I for one hope you act much more seriously on here then you did on paranormalia. However, from what I am reading I doubt you will do that. I am already seeing the pettiness that tends to be typical of your arguments. For example it does not matter if I call it the excluded middle or a false dichotomy, it is the same thing. Your attempt to critique me on that is typical of your style, attack everything but the point of your opposition. You use red herrings to attempt to distract from serious discussion. That is why I call you a troll.

I have on numerous occasions pointed out your poor knowledge of what you seek to critique and your blatant fallacies , your favorite by far being the excluded middle ( I see no need to pander to you by calling it a false dichotomy), but certainly it is not your only fallacy of choice. (special pleading is certainly another one you like)

Seriously everyone in here I am putting the troll alert on Harley. Go to paranormalia and read the following posts and see Harley's behavior.

'Horizon' on dogs

Dawkins on Haiti


The Twin Thing

If you read his responses to myself, the Major and Paul you will see a man who deliberately misrepresents people to make a point.

"Michael – I am wondering what you mean by 'statistically impossible'"

Consider a fairly well-known calculation made by information theorist Hubert Yockey. He determined that, given a quantity of 10 to the power of 44 amino acids combining and recombining in solution, it would take 10 to the power of 23 years to produce (by chance) a particular protein, cytochrome C, which is one of the building blocks of a cell. Since the universe is believed to be only 14 billion years old (one trillionth of the time period required), there would not be enough time in all of history to produce this protein by chance. And since a cell requires many proteins, the origin of a cell by chance becomes so unlikely as to be, for all practical purposes, impossible.

Of course, it's possible to challenge Yockey's assumptions. For instance, there may be other paths to a protein that would be functionally equivalent to cytochrome C. This web page makes that argument:

Regardless of the merits of Yockey's conclusion, the point is that some events are so extraordinarily unlikely to occur by chance that chance is effectively ruled out. I would say these events are "statistically impossible" (by chance).

Whether or not this is true of multiple lottery winners, I have no idea.

Ya'll's banter about biochemistry reminds me of this quote by James Jeans,

"Life exists in the universe only because the carbon atom possesses certain exceptional properties."

I'm an "everything is a miracle" kind of guy.

Michael – I’m not an evolutionary biologist, nor am I familiar with Hubert Yockey, but I don’t actually disagree with what you report he says. And there is also the standard tornado-in-a-junkyard-creating-a-jumbo-jet argument. Even if the junkyard contained only standard parts for a 747 rather than assorted rusting cars and trucks, I don’t think there is a reasonable chance that an airliner of any type (or anything else you can think of) is going to come together within the lifetime of a million universes. With evolution, it seems that something more than just chance is at work, but I would go with the fact of KNOWN chemical reactions building upon each other over billions of years rather than the intervention of a deity or “intelligent designer,” for which there is no objective evidence.

With regard to the multiple lottery winners, I remember something similar in my home town quite a few years ago. It involved bingo rather than the lottery, though. It turned out that a couple of winners in the biggest games (where many bingo halls link together to pool the winnings in one huge pot) had an extraordinary run of big wins. As I suggested in my earlier post, the bingo companies noted the fantastic odds involved, but did not assume psychic intervention; the odds against such a win were, obviously, astronomical; the company instituted an investigation because their hypothesis was that fraud was a possible factor. And that, indeed, turned out to be the case. Several people were jailed for a fraud that was simple in its application, but ingenious in its conception (essentially, someone spotted a loophole in the system, and had the nerve to exploit it in full view of everyone, none of whom had even thought of it).

Is it possible to defraud the lottery, with all of its security procedures? It would be difficult, to be sure, but I don’t think the possibility can be discounted. It’s unlikely, but the possibility that a lucky winner then goes on to spend a fortune on more and more tickets is also a possibility. A few people I know seem to have very good luck with raffles and other games of chance, but I know they also buy many entries to those games, and probably end up out of pocket by the time the value of their winnings is balanced against the money they have spent to be able to win.

The bottom line here for me, though, is this: when something highly improbable happens, why should a psychic component be invoked rather than just chance or even outright fraud? [Note to Kris: have I violated Aristotle’s Law of the Excluded Middle there?] Any of those three possibilities might be true (and there might even be other possibilities [take note, Kris] that I have not thought of). A hypothesis of fraud (as I have just mentioned in the bingo example) was investigated and proven to be true. If a hypothesis of psychic intervention can be investigated and be proven to be true, then I will go along with it. Do I sound unreasonable with that?

No, Harley, you don't sound unreasonable. As I said, I threw out this idea as a fun piece of speculation. I don't imagine it would convince anyone of the reality of psi. But if psi is a reality (as I think it is, for reasons having nothing to do with lotteries), then we might expect it to play a role in some games of chance.

With regard to the origin of life, one thing to keep in mind is that, if life originated on Earth and did not arrive here on a meteorite, then there weren't billions of years available for the first living cell to develop. The time period between the cooling of the Earth and the first appearance of one-celled lifeforms was less than a billion years - possibly a lot less, depending on exactly when Earth cooled sufficiently (between 4.4 and 4.1 billion years ago) and exactly when the first life arose (3.5 billion years ago at the latest, but possibly 3.85 billion years ago or earlier).

My personal view is that there are three levels of reality: 1) physical reality, 2) an informational substrate that generates physical reality more or less the way a computer's information processing generates images on a computer screen, and 3) a universal consciousness or Mind, which "wrote the program" that is being executed on level 2. If there is any truth in this, then it might not come as a surprise that living cells are based on encoded information (DNA, RNA). After all, in this model everything physical develops out of information.

Hi Harley, I agree with you that the case with multiple lottery winners can't be used as an argument for or against psi, even though the a priori probability for winning the 4 lotteries are approx. zero.

I'm interested in your thoughts on this article below - where a scientist have used standard statistical methodology to prove a psi effect. As being trained in statistical analysis you will have to accept the mathematical significance of the study.

As being a completely new experiment it's not possible to accuse the author for publication bias as the file drawer effect isn't in play as with conventional psi research. (but ofcourse you can accuse the author of outright fraud).

Michael – speculation is fine by me. I imagine the greatest scientific discoveries must have started as speculation.

No-one knows exactly how life originated, but some biologists “speculate” that it could have started with a simple self-replicating molecule; but although I am not a biologist, I don’t think evolution is just about billions of years, but how many generations are involved. Simple chemical reactions can happen very quickly, so I suppose there could be millions of generations per hour right at the very beginning. But that is just my own speculation. I look at the work of people like Craig Venter with some optimism.

I agree with your first idea of reality, i.e. that there is an objective reality out there (notwithstanding unproven Cartesian demons); but your points 2 and 3 really are speculative. I’ll have to think about that.

sbu – thanks for the link. I’m following the story and looking at other websites and blogs that are taking an interest.

I would be cautious, though. It’s OK by me that this new research is going to be published (proving, perhaps, that there is not really an institutionalised suppression of this kind of research). There are further hurdles, not least the fact that replication is a necessary part of the process before it is going to be accepted.

I would say this, however: even if there is something remarkable going on, Bem has not proven the existence of psi. It is ONE possibility among many. And as far as the statistical significance goes, it is not huge, and could be due to chance variation; we will have to see what happens when attempts at replication get into full swing (attempts have started, but it’s too early yet to know what will happen).

I’m not going to accuse the author of fraud, although the history of psi research has its fair share of it. I’m happy to regard this as genuine research, but, like all scientific research, it will take some time before a conclusion can be reached. As you will know, scientific points are never settled on the basis of a single study or experiment. Personally, I don’t think this is going to be the big breakthrough, but time will tell.

Bem's experiments are not the first ones of this type. Dean Radin performed similar tests, which he discusses in his book "Entangled Minds." The results he obtained were mixed, but in his first series of tests he did find that the "presentiment effect" occurred "with odds against chance of 500 to 1." (p. 166) He conducted several such experiments. One participant is quoted:

"It's spooky. You sit there and watch this little trace, and about three seconds, on average, before the picture comes on, you have a little response in your skin conductivity which is in the same direction that a large response occurs after you see the picture.... He's done that over and over again with people." (p. 170)

So although the story is new to the media, it's not really new in psi circles.

Radin comments on Bem's paper here (see the comments thread):

"I’m not going to accuse the author of fraud, although the history of psi research has its fair share of it."

Harley, did you mean to imply that psi research has historically had a greater problem with fraud than mainstream scientific research? If you did—and it's hard to tell from the way you phrased it—I don't think that's accurate.

Harley, I have several issues with the idea of lab controlled, replicable psi research.

First, to me, psi is more like the weather. It is probably, at least in part, the result of a host of random (or chaotic) variables coming together in just the right way at just the right time; though there may be some involvement of conscious intent as well.

No one doubts that Hurricane Katrina occurred, but go ahead and try to replicate it. Try to even accurately predict another occurence within the range of probabilities that you consider to be suggestive of proof.

But, more practically, the whole notion of random selection of subjects in psi experiments smacks of being ill-conceived. It seems reasonable to think that some people (a minority) are going to be better at psi than others (the majority?) and that even those who tend to be better at psi will have off days.

There may even be people who block their psi abilities due to some prejudice and/or fear. The unskilled and the "blockers" would drag the statistical significance down, right?

I would prefer to see studies where a large pool of subjects is started and tested and the pool is whittled down to a meaningful sample that has demonstrated the highest psi rates. In other words, ultimately only study only subjects who are proven adepts. I think this just might increase the statistical significance of the results. And I think it is scientifically reasonable.

The psi experiments I read about are somewhat analogous to this; I claim men can benchpress 600 hundred pounds. You've never seen this before and are skeptical (probably citing all sorts of bio-mechanical authorities).

To test the claim, we randomly select a groups of men from college psychology classes......well you can guess that your null hypothesis will stand.

However, if we were to select from the population of men working out at Muscle Beach (in Venice, CA) I'm pretty sure we would see a statistically significant effect in favor of my claim.

Hopefully you get the point. I'm surprised that serious experimenters like Dean Radon don't take this into consideration as well.

Oooooops... I spoke too quickly......apparently Dean Radon has considered increasing effect by using skilled subjects.

"I would prefer to see studies where a large pool of subjects is started and tested and the pool is whittled down to a meaningful sample that has demonstrated the highest psi rates."

It's a good point, no one, but isn't it true that many psi studies are, in effect, carried out in such a way that scientists end up focusing on those who seem to be the most gifted? I'm thinking, for example, of Targ and Puthoff's remote viewers, Stephan Schwartz's experiments in the same area, Gary Schwartz's mediums, even Sheldrake's work with Jaytee, the psychic dog.

Yes, Bruce, you're right. There is a little of that here and there, but, overwhelmingly, the studies and meta-analyses that are cited as proof of psi (and that are debated by skeptics) typically involve ordinary subjects.

This is unfortunate because the resulting psi effects - though real - are typically very small.

Some of Targ's subjects made remarkable "hits" in the remote viewing experiments. So remarkable that one pretty much has to either accept the reality of psi or call Targ a liar.

I think the whole problem with identifying a group of adepts and running them through the experiment is that replication becomes, practically speaking, difficult. In order to replicate one would either have to bring back together the same group of subjects (very difficult and expensive; especially if the experiment is to be replicated in a different geographic location) or go through the process of finding a whole new group of adepts; a process that would greatly increase the time and resources involved in the study.

By generally not investing in identifying and focussing research on adepts, psi researchers have shot themselves in the foot and have retarded the field of study.

I've had quite a few mystical experiences, precognitive dreams, telepathy with my wife, synchronicities, etc. I even heard a voice one time that told me what my wife was fixing to do a minute later but I've never been able to control it. They just happen when they happen. It's definitely not me making it happen because if it was I'd do it all the time because I like it when it happens. When it does happen to me I'm like "WOW" that was amazing but I've never been able to make it happen. The other side doesn't seem too keen on being anybody's servant or being at anybody's beck and call.

There is a little of that here and there, but, overwhelmingly, the studies and meta-analyses that are cited as proof of psi (and that are debated by skeptics) typically involve ordinary subjects.

There is quite a "who came first - the chicken or the egg" issue to your proposal. Imagine a paper issued to a scientific journal with an abstract saying "In order to prove psi we use subjects with proven psi ability....hmmm"

I think researchers gravitated toward ordinary subjects to address concerns about cheating. If you test, say, Uri Geller and get good results, critics will inevitably say that Geller fooled you with sleight of hand or other tricks. The same is true if you test a professional medium, who can be accused of being a mentalist. But a person off the street is unlikely to have either the motivation or the skills to successfully cheat, and a random sample of hundreds of people reduces the risk of cheating even more.

The trade-off, though, is that ordinary folks are unlikely to have robust psi abilities, so the results will show only statistical significance at best.

Personally I think both approaches have value, but perhaps more emphasis should be put on people with exceptional psi abilities, as was done in the early years of psychical research.

"In order to prove psi we use subjects with proven psi ability....hmmm"

Suppose it was phrased this way: "In order to prove robust psi ability in exceptional individuals, we first identified subjects who appeared to demonstrate this level of ability in more informal tests, and then subjected them to a series of increasingly refined tests ..."

Michael – one very good thing about Bem’s experiment is that he has provided open access to his data and the software he used, so he is not going to be accused of hiding anything. That is how good science should be. On the other hand, there has been at least one failed attempt at replication, and his scoring method has been criticised. It’s still early, though, so I’m not going to go on the offensive, as it were.

Bruce Siegel – mainstream science also has its “fair” share of fraud, and I would not try to defend any scientist who has engaged in any kind of dishonesty. On the contrary, when a scientist has been caught in fraud or any kind of dishonesty, then that person should be dealt with severely. A recent and very high profile example is the case of Dr Andrew Wakefield, who has now been struck off the UK medical register. He gets no sympathy from me, or anyone else for whom the integrity of science is a top priority.

no one – not everything in science is done in the same way. Geology cannot be subjected to randomised double blind control group experiments, and nor can a lot of other areas of science, but they are still science, investigated by other methodologies.

With regard to hurricanes, the study of them is replicated every time one occurs. Obviously, Katrina itself is now finished, but all hurricanes follow a predictable development and the study of each new one is, in essence, a replication of the previous study. Although no-one is going to be able to predict the date, time and location of a particular hurricane in the future, the mechanisms are fairly well understood (but that’s not to say that there aren’t things yet to be discovered about them). In fact, science has a predictive power that is simply unmatched by any proclaimed psychic. When a hurricane starts to form, I think people consult the weather forecast rather than a psychic.

I agree that those who claim psychic abilities are those who should be tested. I especially agree if they would allow themselves to be tested in the same open manner as Bem has done with this latest research.

Michael – you implied in the earlier part of this thread that some lottery winners, for example, might have psychic abilities they are not aware of. If so, then I would go along with the idea of testing randomised samples of the population. It would be interesting to see if, over a very large sample, there are those who appear to have psychic abilities they don’t know of, who also happen to be the people who have an unusually high winning rate in games of chance.

But I would also refer to your opening remarks, “ just doesn’t work like that.” No, alas, it never does.

The comments to this entry are closed.