[Chris Noble]

Originally Posted by David Bailey

Chris, I rather think you have never done any actual science, because if you had, you would realise that hardly anyone does research without some idea of what they want to demonstrate - it just doesn't work that way!

As usual your thinking is completely wrong.

Obviously, his experiments are meant to eliminate explanations 6-10. Explanations 2-5 are difficult to tease apart, and probably overlap one another. In any case, any of 2-5 would have enormous significance. Not even you accept explanation 1, I think.

That's the main thrust of what I am saying. I think Sheldrake is entitled to reject hypothesis 1. This does not entitle him to choose any of the other hypotheses as being proven.

Sheldrake designed his experiments so that if there was any sensory leakage whatsoever then the dog would on average spend more time at the window during the homecoming than in the main period. The experiments are designed to get an effect but not to explain it.

My point is that it would be far more useful to make the next experiment more effective at eliminating explanations 6-10, and if you think you can make useful suggestions to Alex, why not go ahead? For the most part, science advances by doing more experiments - not by endlessly criticising those that have been done.

I just made some useful suggestions. The benefit of testing hypotheses 6-10 individually is that the contribution of each can be quantified and this can lead to better controlled experiments in the future.

As I see it, the problem with at least some animal behaviour research, is that it never even considers ? explanations - so it does not try to prove them, or rule them out. However, to rule them out, they would have to do an experiment that may well achieve the opposite.

The problem is that the only definition of "psi" is "not explained by natrural science". The only way of testing "psi" is to investigate natural science. Think of animal navigation. How can you possibly test psi explanations? The only thing you can do is investigate natural explanations.

Forget 'morphic resonance' - I would concede that this concept is too vague to be considered a real theory. However, again you don't seem to realise the way science operates in speculative areas. Often researchers don't have the luxury of a number of well developed theories to choose between.

Stop trying to tell me that I don't understand how science works. Even Einsteins theories of relativity built upon existing knowledge. He didn't just pluck his ideas out of thin air.

Remember. hardly any experiment is completely beyond criticism. Indeed, scientific papers often end with a section outlining possible improvements that could be incorporated into future experiments. Sheldrake's experiments to determine if a dog can be aware of the randomly determined return time of the owner seems to me to be designed to rule out explanations 6-10. Maybe they can be improved upon, but his concept is certainly sound.

Sheldrake only pays lip service to the various natural explanations for the dog's behaviour so that he can ignore them. He never explicitly tests the assumption that his experiments control for these factors. He also assumes that any natural explanation has to explain the entire effect

I would say that the most effective approach is not to try to rule out 6-10 separately, but simultaneously. The random return times should eliminate 6,8,9,and 10. Furthermore, I understand that at least some of the trials used a taxi to return home, which more or less eliminates explanation 7.

The effective approach is not just to attempt to rule these factors out but to test to see whether you have done so. Even the supposedly random return times were not very random. The dog only needed rough information about the approximate return period in order to reproduce the observed pattern of being at the window mnore often in the return period than the main period. Why the reluctance to actually test this properly. Do one set of trials with the owner returning at routine times of her own choosing and do another set where the owner returns at truly random times over a 24 hour period. The null hypothesis would be that there will be no statistical difference in the amount of time that the dog spends at the window during the owner's return between the two groups.

When we returned home, our cat would ask to be let out of our neighbour's house ahead of our arrival - that must be a good minute before we arrived home in a pretty standard make of car. That seems to make explanation 7 pretty unlikely. At the time, I had exactly your mindset, so I laughed in a friendly but disbelieving sort of way - now I wish I had taken the opportunity to do a few experiments!

Part of the normal etiquette of science is that before you criticise an experiment you check with the experimenter. Lots of common sense experimental details never make it to the paper, so for example it is no use reading a paper and screaming "I know why this gave an odd result - they don't say they washed their test tubes before they started!". Likewise, I suspect if you contacted Sheldrake about these experiments, you might be reassured (if that is the right word) that he was far more thorough than you think.

Despite your assertions I am very much aware of the way in which normal science works. I know that if I am going to stand up at a conference and claim to have discovered a radical new theory that completely overturns modern science that I had better make sure I have covered all possible criticisms of my work. If I present shoddy, amateurish work then I would expect to be strongly criticised. Anyobody who has submitted a paper to a journal knows only too well the criticism that is an entirely normal part of science.

I don't have to provide the explanation for Sheldrake's results. He has to demonstrate, rather than just assume, that he has controlled for the perfectly natural explanations for the phenomenon.

I really don't think that Sheldrake is interested in finding explanations for the phenomenon. He already has his explanation - "morphic resonance". If he were really interested in the phenomenon he would have done more research to tease out the details. As it is he simply stopped when he had some preliminary results that appear to support his claims.

At times you seem close to saying that 'science' should not even try to research claims of ? - just concentrate on pumping out propaganda telling everyone it is bunk!

At times you seem to want to put words in my mouth. If you can come up with a theory of psi then investigate it. While "psi" is defined as something inexplicable via current natural science then "psi" is simply anomaly hunting and not science.