TED Conversations

Emily McManus

Editor, TED.com, TED

TEDCRED 200+

This conversation is closed.

Rupert Sheldrake's TEDx talk: Detailing the issues

There's been a lot of heat today about Rupert Sheldrake's TEDx Talk. And in the spirit of radical openness, I'd like to bring the community into our process.

http://www.youtube.com/watch?v=kO4-9l8IWFQ

While TED does not vet speakers at independent TEDx events, a TEDx talk can be removed from the TEDx archive if the ideas contained in it are wrong to the point of being unscientific, and that includes misrepresenting the scientific process itself.

Sheldrake is on that line, to some commenters around Twitter and the web. His talk describes a vision of science made up of hard, unexamined constants. It's a philosophical talk that raises general questions about how we view science, and what role we expect it to play.

When my team and I debate whether to take action on a TEDx talk, we think deeply about the implications of our decision -- and aim to provide the TEDx host with as clear-cut and unbiased a view as possible.

You are invited, if you like, to weigh in today and tomorrow with your thoughts on this talk. We'll be gathering the commentary into a couple of categories for discussion:

1. Philosophy. Is the basis of his argument sound -- does science really operate the way Sheldrake suggests it does? Are his conclusions drawn from factual premises?

2. Factual error. (As an example, Sheldrake says that governments do not fund research into complementary medicine. Here are the US figures on NIH investment in complementary and alternative medicine 2009-2010: http://nccam.nih.gov/about/budget/institute-center.htm )

As a note: Please know that whether or not you have time or energy to contribute here, the talk is also under review by the TED team. We're not requiring your volunteer labor -- but we truly welcome your input. And we're grateful to those who've written about this talk in other forums, including but not limited to Jerry Coyne, PZ Myers, Kylie Sturgess and some thoughtful Redditors.

Share:

Showing single comment thread. View the full conversation.

  • Mar 7 2013: As a graduate student in computer science and artificial intelligence and as someone who frequently reads stuff in biological vision, Sheldrake seems to be inhabiting some scientific world I have never ever seen. Most of what he claims about the process of science is just decidedly untrue. A lot of what he calls "dogma" is just a working hypothesis in science that has served us well, and for which no real plausible alternatives exist. Consider for instance the "dogma" that "memories are stored in the brain". Contrary to what Sheldrake implies, we have several decades of computational models of how memories can in fact be stored in neurons (see perceptrons, multilayer perceptrons, Kohonen maps, associative memory, and a whole plethora of very recent and exciting works on artificial neural networks). Contrary to this, I have never seen a plausible alternative, except vague, undefined words in insubstantial talks (eg. "morphic resonance" or whatever). Simply coming up with a term that sounds good is not the same as coming up with a well-defined, testable hypothesis.
    The part about natural constants is hilarious and should embarrass even high school science students. In his whole rant about how the constants might change, he never even considers the possibility that measurements have noise, and uncertainty improves as better techniques are discovered. He says the value of c decreased: by how much? Was it within the error bars?
    In addition, the reason why these are constants is not because someone decreed them to be constants, but because there are theories that predict that these are constants. G is the proportionality constant in Newton's law of gravitation. To say that G is not constant is to question the validity of the law, a law which actually *predicted* the existence of Neptune before it was discovered. Not many laws can predict the existence of whole undiscovered planets. I would like Sheldrake to come up with an alternative that has similar predictive power.
    • Mar 7 2013: It should be noted, in support of your statement, that error bars only cover the precision of data, not its accuracy. One can have very consistent data, but have systematic errors that compromise accuracy of the data. I can measure my room very precisely by repeated use of a worn-out/stretched-out tape measure. The data, if I have collected many measurements, would have very small error bars and but the accuracy would be off, my room would appear smaller than it really was.

      Past methods for measuring the speed of light were both less precise (larger error bars) and had systematic problems (as all measurements do to some greater or lesser extent) which led to decreased accuracy. Sheldrake has interpreted the minimizing of systematic errors over time as real data. While there may be something there, there is no way to tell, as it is lost in the "noise."
      • Mar 7 2013: True, and I don't know how the accuracy has changed over the years. I mentioned precision because it seemed safe to assume that precision would have improved :).
    • thumb
      Mar 7 2013: Why don't we take Rupert's 10 itemized issues one-at-a-time and, using the Rules of Logic, prove each one to be false? If he is 100% in error then, once corrected, they should read as follows: 1) The Universe is purely mechanical. There is no soul, or spirit in Man. 2) Consciousness has no relationship to material things and vice-versa. 3) Natural Laws and Physical Constants can never change. Once "they" define a Law/Constant it is impossible for it to change 4) The sum of matter/energy in the Universe can never change. 5) There is no purpose associated with the Universe. It may continue forever for no reason, or it may come to an end for no particular purpose. 6) Inherited characteristics are a result of material/chemical actions. 7) Recalling stored information from the sub-conscious mind to the conscious mind is a purely material/chemical function. 8) The act of conscious thought is solely dependent upon the brain. Like a beating heart outside the chest cavity a brain outside the skull could be conscious of its condition. 9) Thoughts cannot act through a distance because they are confined to the brain. 10) Biological ailments can only be corrected using medicine. A healing without medicine is simply a spontaneous cessation of the cause. Are those all justified true beliefs? If so, the result of the exercise is that we understand better what we claim to believe. Should everyone who disagrees and challenges us be silenced? I don't think so. Publish the Talk!
      • Mar 7 2013: Two things here, before we do that:
        1) Logic is not what we want, or rather not the only thing we want. Logic is sufficient for math; for science we need something more: notions of evidence, and notions of uncertainty.
        2) And before we do any arguing or debating, we need to convert any terms which we might loosely throw, into well defined, testable, hypothesis. By that I mean that I should take a statement like "The universe is mechanical" and decide what predictions such a statement makes that *can be subjected to an experiment, and so falsified*.
        Which is why we need to be careful about statements like "The Universe is purely mechanical", or "Consciousness has no relationship to material things". What exactly does material mean here? Is the electromagnetic field material? Is a computer program material? What does mechanical here mean? What exactly does consciousness mean?etc.
        Now, I personally would be happy to talk about the validity of tens or hundreds of theories, but only *once they are fully specified*. Notwithstanding Sheldrake's many factually incorrect statements, he is also woefully vague, using ill-defined terms that can have any number of meanings in lay usage. That, to me, is not scientific, and is particularly out of place in a talk about science, of all things. Present alternative views all you want, but at least do so in terms of words and terms that can be unambiguously understood by all scientists who come to the table.
        In other words, people who disagree must not be silenced, but this does not automatically make every person who has not even made the effort of defining his views properly worthy of being listened to. Given the limited number of talks that TED has, I would much rather listen to someone I can learn something from, rather than just a flurry of senseless English words embedded in laughable inaccuracies and misunderstandings.
        • thumb
          Mar 8 2013: Agreed sir, in fact that is my motive for item-by-item falsification of Mr. Sheldrake's talking points. I guess the problem here is that we are arguing about the value of the content of the talk when we should be talking about the refusal to publish the talk. I understand, but disagree with the syllogism that says: TED Talks are factual. Mr. Sheldrake's talk is not factual. Therefor it should not be a published TED Talk. The second premise is the problem. Determining the factual fidelity of the talk is way beyond the scope of this forum. Each of us must make that determination on our own. To abort a person's right to participate in the premier online discussion forum because they are less than perfect regarding Truth is not a good idea. If - I said IF- the post, or Talk, is crisp, candid, provocative and relevant then publish it!
        • Mar 8 2013: Sheldrake's talk is definitely unsuitable for TED as it it not based on fact or logic. However, rejecting him from this forum does not silence him. After all he has published all these silly notions in a book already. The contents of the book would not be accepted by a serious scientific publication and I don't think TED should dirty it's hands with it either. The book will probably be read by more people than know TED exists.
      • Mar 7 2013: Just to add, I am not saying we should outright reject contrary views. I am just saying that people talking on TED or TEDx should have done the effort of defining their views properly and collecting evidence for and against them.
        • thumb
          Mar 8 2013: I agree that a polemic is not the best way to make the points Mr. (Dr.?) Sheldrake attempts to make. But a poor choice of style is no reson for being banished.
    • Mar 7 2013: "The part about natural constants is hilarious and should embarrass even high school science students. In his whole rant about how the constants might change, he never even considers the possibility that measurements have noise, and uncertainty improves as better techniques are discovered. He says the value of c decreased: by how much? Was it within the error bars? "

      I think you may be missing the point. We can only dismiss variations from the expected figure as 'just noise' if we know in advance what the 'right' answer 'should' be. But how did we find out in the first place what these constants are? By experiment and measurement, which always contain noise. See the problem?
      • Mar 7 2013: Not really. You are looking at these things in isolation, which they are not. The speed of light is not the same as the current temperature in Berkeley.

        The speed of light for instance is one part of a large coherent scientific theory: the theory of relativity. The theory of relativity has made many predictions, and has been tested numerous times, and so, provisionally, we accept it as largely correct.

        Since this theory is accepted as largely correct, and this theory accords a special place to the speed of light as a constant, we accept, provisionally that the speed of light is a constant.

        Now, given this, we see that measurements of speed of light vary. Which is more likely, that the theory is wrong, and so all of its myriad accurate predictions are completely wrong and all of astrophysics which relies on this must go out the window? Or that grad students tend to press stopwatch buttons a bit too soon?

        There are two parts to science: experiment and theory, each with uncertainties, each with noise. Weighing all these uncertainties together is not easy. To say that something is possible is trivial, and everything is possible; the question is, which explanation is more likely?

        Caveat: the last time I read about the speed of light was in high school. I literally am at best a high school student in physics.
        • Mar 7 2013: Thanks for your thoughtful reply Bharath.

          I assure you I am not looking at constants in isolation. My apologies if my answer gave that impression. On the contrary, I was trying to give a simple example of the method of scientific inquiry.

          We never observe "the speed of light". We observe measurement1, measurement2, measurement3, etc. Agreed?

          What about the "large coherent scientific theory" that goes with it? Do we observe that directly? I don't see how we could (and no philosopher I ever read says we do).

          Where does the "large coherent scientific theory" come from? From the same place all scientific theories do: the human imagination interpreting measurements/observations. Agreed?

          There's a doctrine many (but not all) scientists believe: measurement1, measurement2, measurement3, etc. prove the existence of The Constant, and likewise, evidence1, evidence2, evidence3, prove the existence of The Law.

          Is a testable scientific hypothesis or a philosophical assumption? Take a moment and consider that.

          This is the 3rd dogma Sheldrake refers to. It's Platonism plain and simple.

          I am not smart enough to see how we can believe the material universe is governed by pure, perfect laws and constants without positing an immaterial, cosmic layer to the cake. I'm not saying that the Platonic realm, scientific constants, scientific laws or the Invisible Pink Unicorn definitely don't exist - just that no one's ever seen them. And even if we do accept that there are these Laws, that doesn't solve our problem: where do these laws come from? Why do they exist? Where do they exist? How is it possible to have an material universe governed by immaterial laws?

          So Sheldrake's offering a different explanation for why our observations are kinda sorta fairly consistent most of the time: universe is habitual, and the more something happens, the more it will continue to happen. This is the broader scope of the morphic resonance theory as I undertstand it.
      • Mar 8 2013: Okay, I am sorry I didn't get that point; I get it now. To be frank, there is no way I could have got this from Sheldrake's talk.

        My answer to your comment will have to be that I don't know. I don't know where these laws come from. And it is an assumption that such laws exist, and such constants exist.

        But here's the thing. That assumption has helped us a long way. That assumption works. The things we predict about astrophysics come true to a large extent. The things we predict about physics and chemistry and biology come true to a large extent, enough for us to build machines and spaceships, treat diseases, understand a vast number of natural phenomena.

        These assumptions are not dogma. It is certainly possible that these assumptions are wrong. Maybe a truer alternative exists. But given the success of what my current assumptions are, for me to think about a particular alternative, someone will have to 1) define the alternative precisely, and 2) show that the alternative not just explains all, or most, of what I have managed to explain till now, but more.

        Sheldrake on the other hand simply handwaves at vague words that make no sense to me, to state it simply.
        • Mar 8 2013: "To be frank, there is no way I could have got this from Sheldrake's talk."
          True. I did my best to represent my understanding of Sheldrake's theories, based on what I've read of his works, not just the talk. He uses the phrase "cosmic Darwinism" somewhere; the consistent patterns in universe's behavior evolved.

          "But here's the thing. That assumption has helped us a long way. That assumption works. "
          But alternatives like Copenhagnism or model agnosticism work equally well if not better.
      • Mar 8 2013: "But alternatives like Copenhagnism or model agnosticism work equally well if not better."

        I'll have to read up on that; thanks!

Showing single comment thread. View the full conversation.