Dec 09 2008

SETI Science

Published by under Uncategorized
Comments: 30

I am a strong supporter of SETI – the search for extraterrestrial intelligence. To me this is a fascinating scientific endeavor with a potentially huge payoff. And yet, I find that even within skeptical circles I hear grumblings that SETI is not real science.

The primary complaint stems from the misapplication of an important scientific principle – that a necessary criterion for any scientific hypothesis is that it needs to be falsifiable. If you cannot make an observation or conduct an experiment to prove an idea wrong, then that idea is not useful to science.

SETI critics argue that the notion of extraterrestrial intelligence is inherently unfalsifiable. No matter how much you look with negative results, one could always argue that it is not enough, we have to look deeper into space, for fainter signals, in more EM frequencies.

But this, however, does not render SETI unfalsifiable in the that is meant by philosophers of science, because it runs up against another scientific principle – the notion that you cannot prove a negative.  It is logically impossible to prove completely that a phenomenon does not exist (unless it can be demonstrated that it violates know laws – but even then the proof is only as good as the degree to which the violated laws are established). Evidence for the lack of a phenomenon is always only as good as the thoroughness and efficacy of the search methods.

You can apply this to anything is science, even drug trials. If a large well-designed clinical trial shows that drug A does not improve disease B what it really showed is at the dose and duration given it did not have an effect big enough to be detected with statistical significant by the study. You cannot rule out a small effect, a subpopulation effect, or that a higher dose would not have shown an effect. What the study does, in effect, is set statistical limits on the size and scope of an effect.

The search for ET (or any new phenomenon) is the same.  We can never prove that there are no ET’s in the universe – but that’s not the point. The more searching we do with negative results the more statistical limits we can set on the number of ET civilizations that are engaging in specific behaviors – like broadcasting at certain frequencies. Right now, for example, we can say there aren’t numerous nearby ET civilzations or artifacts braodcasting in our direction in the radio frequencies we have listened to.

In a recent Nature commentary, Philip Ball also feels the need to defend SETI.  He discusses another method (other than listening for radio signals) for looking for ET, writing:

Dyson suggested that a sufficiently advanced civilization would baulk at the prospect of its star’s energy radiating uselessly into space. They could capture it, he said, by breaking up other planets in their solar system into rubble that would form a spherical shell — known as a Dyson sphere — around the star, creating a surface on which the solar energy could be harvested.

He is talking about the work of Richard Carrigan, who asked a specific question: are there any detectable Dyson spheres within about 1000 light years. That is not the same thing as asking if there are any ET’s, rather its a proxy question. It’s a method of looking where the light is good. What behaviors might ET be engaged in that we can detect. Well, they may be broadcasting at us in the radio frequency. Or they may have buit something like a Dyson sphere around their star – perhaps we can detect that.

Of course, these methods require speculation about what an ET civilization might do. This is problematic because really we have no idea. All we really can do is assume that humanity is somewhat representative of technological intelligence and then speculate about what a more advanced version of our own civilization might do. It’s a pretty thin line of reasoning, although it is legitimate as far as it goes. We just know we are speculating from too little information.

Another recent SETI “innovation” was to assume that ET might be interested in looking for life elsewhere in the galaxy. They might start by looking for planets around other systems, and planets like earth would be easiest to detect from a direction in the plane of our system. Therefore, to those observers the earth would eclipse the sun and that is one way they could detect it. It might be more likely, therefore, that such civilizations are broadcasting at us, because they know we are here, or at least life might be here. Perhaps, then, SETI should focus its efforts by looking at stars in the plane of our system.

That is another long chain of speculation – but very interesting, and legitimate as far as it goes. This is an effective investigational logic to use – imagine what might be the case, then take a look.

It still strikes me as odd when people use this feature of SETI to argue that it is pseudoscience. No -it would only be pseudoscience if SETI proponents concluded that ET exists based on such reasoning, or to explain away negative results. But this is a perfectly reasonable hypothesis upon which to design a program for searching for data.

SETI programs also use a very reasonable protocol for dealing with signals when they encounter them. They consider such signals an anomaly, then they systematically rule out possible known causes.  So far, every candidate signal has turned out to be something known – no enduring anomalies. This behavior is very different from pseudoscientists, who might call such signals “ET signals” and then proceed to only rule out a couple of obvious known causes, or then dismiss non-ET explanations for the signals.

Take ghost hunters, for example.They typically look for anomalous photographic effects, declare such anomalies to be “ghost globules” and then dismiss any non-ghost explanation.

I think it is important to properly demarcate legitimate science from the “cheap imitation.” In my book, SETI is legit.

And, of course, with SETI we just might hit the cosmic lottery and find genuine evidence for ET. That’s worth the tiny investment currently going to SETI projects. Even knowing that our galactic neighborhood is relatively empty is interesting information. Perhaps humanity is more rare than we imagine.

But of course nothing would be more fascinating than discovering that we are being bathed in a stream of information from an advanced alien culture.  I like just knowing that SETI makes that possible.

30 responses so far

30 Responses to “SETI Science”

  1. caoimhon 09 Dec 2008 at 9:34 am

    I couldn’t agree with you more Steve.
    The SETI institute is one of the areas of science that inspires me greatly.
    Seth Shostak (SETI Astronomer) presents a great podcast too called “Are We Alone”. I highly recommend it to SGU viewers. It’s right up your alley!

  2. Steven Novellaon 09 Dec 2008 at 10:29 am

    We interviewed Seth here:

    He is great. He also does a great job communicating during interviews, for example on his recent appearance on Larry King opposite some UFO nuts.

  3. alfon 09 Dec 2008 at 11:09 am

    Good article, Steve.

    One way you can think about the “falsifiability” objection is to turn it on its head.

    If the hypothesis is: “There is intelligent life in the universe other than humans,” then no amount of evidence can prove this wrong.

    But, if the hypothesis is: “There is NO intelligent life in the universe other than humans,” then you just need a single example to disprove the hypothesis.

    It seems like SETI is assuming the latter, and doing their best to disprove it. That’s good science.

  4. derekjameson 09 Dec 2008 at 11:14 am

    I don’t happen to agree with the falsifiability criterion for good hypotheses, but I do think SETI is highly speculative. So I think it’s appropriate that it’s funded privately…I wouldn’t support tax money going to SETI, but I applaud their efforts and wish them luck.

    The main scientific issue I see with SETI is that it involves a search for the conjunction of two events (the origin of life and the origin of human-level intelligence) that we know very little what the probability of occurrence is. Good work is going into trying to understand abiogenesis, but right now the issue is very poorly understood, so we can’t make any solid estimates about the likelihood that it or a similar event would occur somewhere other than earth. Same with the origin of intelligence. So while I wouldn’t equate SETI with a search for leprechauns or ghosts, I also wouldn’t personally invest time or resources into it.

  5. Mozglubovon 09 Dec 2008 at 11:32 am

    Also, I am predisposed to liking the SETI program because it was one of the best wonders in the original Civilizations game…

  6. Jim Shaveron 09 Dec 2008 at 11:56 am

    The search for SETI (or any new phenomenon) is the same.

    Actually, the “search for SETI” is pretty straightforward. They’re at, and the SETI Institute is in Mountain View, California. 😀

    I agree that SETI is most certainly real science and that it is worthwhile, especially given the relatively small amount of resources that it uses today. I also agree that Seth Shostak rocks!

  7. Minon 09 Dec 2008 at 1:04 pm

    I don’t have a problem with passive SETI though I do think its very unlikely to have any fruitful output. I do however have a problem with active SETI, where we are actively transmitting signals into space in the hopes of getting some attention. Science should not be engaging in diplomacy and if someone does hear us out there, they won’t necessarily be friendly.

  8. delaneypaon 09 Dec 2008 at 1:37 pm

    You wrote:

    “And, of course, with SETI we just might hit the cosmic lottery and find genuine evidence for ET. That’s worth the tiny investment currently going to SETI projects.”

    Now rephrase it using the woo of your choice, for example:

    “And, of course, with ghost-hunting we just might hit the lottery and find genuine evidence for ghosts. That’s worth the tiny investment currently going to parapsychology.”

    What is the difference? Prior plausibility?

  9. Steven Novellaon 09 Dec 2008 at 3:13 pm


    First, I don’t have a problem with researching low probability hypotheses. I think we should invest in research like a good investment portfolio – 50% or so in high probability incremental or translational research, 40% or so in open ended basic research, and then 10% in highly speculative research. (These numbers are arbitrary and representative – but you get the idea). So a small investment is improbable research is OK. Public funding should be a bit more conservative, and I have absolutely no problem with private funding for anything anyone is interested in.

    My problem with much of paranormal research is the way proponents carry it out and interpret the findings. They tend to over-interpret anomalies and to not adequately rule out known causes. I made a point of stating that SET research does not do that.

    Regarding your specific example, I do think that SETI is more plausible than ghosts – ET technological civilization would not require any change in our understanding of physical laws. The search methods are highly plausible and have a specific rationale. There isn’t any real rationale for anything that passes as ghost hunting.

  10. Jim Shaveron 09 Dec 2008 at 3:22 pm

    delaneypa asked:

    What is the difference [between ghost-hunting and SETI]? Prior plausibility?

    If I may, yes, prior plausibility is one important difference. From what we know about biology, the universe, and the laws of nature, it is arguably highly probable that other life exists in the universe, and that other intelligent, technologically cabable life is likely to exist in our galaxy. On the other hand, the presumed existence of ghosts has no basis whatsoever in science, is accompanied by no legitimate scientific hypothesis, and is clearly traceable to antiquated, superstitious ideas.

    Which brings us to the real crux of the difference: SETI uses principles and methods of science, and is beholden to whatever the data indicate; ghost-hunters don’t know what a scientific method is, and wouldn’t agree to pay attention to negative data even if they could obtain it properly.

  11. Jim Shaveron 09 Dec 2008 at 3:25 pm

    Sorry, Steve. I had not seen your response to delaneypa before posting mine.

  12. cwfongon 09 Dec 2008 at 4:54 pm

    Alf, the problem with the alternate hypotheses you cited is that neither were based on the probability of their confirmation. It’s not probable that one can confirm a certainty, as in the first example, and therefor not necessary to cobble up another improbable proposition to get around that problem.
    SETI assumes neither of these alternatives, as other posters have clearly demonstrated in their discussion of prior plausibility.

  13. Niels Kjaeron 09 Dec 2008 at 7:25 pm

    For many years, I have been involved in searches for hypothetical elementary particles like the infamous Higgs boson. Many of these particles have unspecified production cross sections and their existence cannot as such be falsified. This is not a scientific problem in the sense that scientists just put upper limits on their cross sections. It is normally when evidence for a signal is claimed that controversies arise!

    In the case of SETI the situation is quite similar. SETI can put limits on the density of ETI as a function of the distance to it and the form of EM radiation in question.

    If SETI finds a signal, it should be easy to reproduce the measurement in the same way as other physics experiments, since the signal is not supposed to just go away. This is in contrast to UFO and ghost searches where the “signals” have striking tendencies of disappearing into thin air.

    The unknown prior probabilities in the Drake equation might be relevant for how likely you feel a claimed ETI signal is, but it is the reproducibility that counts.

    My own SETI speculations deals with the question: Would a hypothetical ETI signal mean that we discover ETI or ETI deliberately sending information to us? And if you were ETI: What information would you send to another ETI just coming of technological age to discover it?

  14. DevilsAdvocateon 09 Dec 2008 at 7:45 pm

    As regards the speculative existence of an ETI who may be able to receive and decode our signals and/or send its own signals, an ETI with some level of technological capability, we hold undeniable scientific quality knowledge of a precedent in the universe for this – homo sapiens.

    If ‘we’, why not ‘they’?

    Whether one looks at the SETI project as a hopeless, well-intended effort that is barely ‘scientific’, but unlikely to produce results, or as a fully warranted scientific exploration, I feel it’s one of those occasions where Science gets to lay down all but the minimally necessary cold objectivity, methodology, and efficacy justifications and allow itself (ourselves) to be just a little bit romantic for once.

    I support SETI for the science, but especially for the romance of it.

  15. Eric Thomsonon 09 Dec 2008 at 8:24 pm

    An interesting topic.

    Steve said:
    But this, however, does not render SETI unfalsifiable in the [sense] that is meant by philosophers of science, because it runs up against another scientific principle – the notion that you cannot prove a negative. It is logically impossible to prove completely that a phenomenon does not exist (unless it can be demonstrated that it violates know laws – but even then the proof is only as good as the degree to which the violated laws are established).

    I hate to be pedantic, but this is one of my pet peeves. Logically, it is as easy to prove a negative as a positive (just use modus tonens A–>B, ~B, therefore ~A).

    In science it isn’t logical proof but empirical demonstration that we are typically after. Even then, I can demonstrate that there isn’t a black hole in the middle of my brain. Of course it employs various scientic laws and measurements, so the demonstration is only as good as the background assumptions. But we prove things aren’t a certain way, or don’t exist, all the time in science.

    Further, you could use the same logic to say you can’t prove a positive. I don’t see any logical basis for an asymmetry claim. (To be really picky I could say when you prove a positive (A), you also prove a negative (~ (~A)).)

    I understand, of course, that your point is that it is difficult to demonstrate that certain types of objects do not exist anywhere in the universe, especially if they have some nonnegligible probability of existing given everything else we know (e.g., ETs versus ghosts). But I’d be quite fine if we did away with the incorrect slogan ‘It is impossible to prove a negative.’

  16. cwfongon 09 Dec 2008 at 9:21 pm

    Eric: Can you prove there are no naturally red feathered swans?
    Or put another way, can you prove I am wrong to assume there are?

  17. Niels Kjaeron 09 Dec 2008 at 9:29 pm

    “I hate to be pedantic, but this is one of my pet peeves. Logically, it is as easy to prove a negative as a positive (just use modus tonens A–>B, ~B, therefore ~A). ”

    I theory I would agree, but I can’t. It is not a question of proving if the universe is A or B, it is a question of proving a positive (ETI) or a zero (NETI). The problem lies in the definition of zero.

    There are at least six (silly) explanations why this leads to an asymmetry:

    The pedantic:
    You cannot define NETI before you have described all possible ETI.

    The Bayesian statistics:
    A positive real number has dimension (rank) 1, and the number real 0 has dimension (rank) 0. Therefore the relative prior probabilities for ETI and NETI are undefined. (This is part of the classical clash between Bayesian/Frequentist and statistics which still does not seem to be amenable).

    The mathematical:
    Properties of the real number 0.0 are not defined.

    The computer scientific:
    As long as it has not been demonstrated that P=NP it is impossible to program it. And what can’t be programmed ain’t easy.

    The astrophysics:
    Our universe is expanding so rapidly that it is nearly impossible to discern what is going far away. At large redshifts only galaxies and exotic objects are visible to us.

    The Hawking’s style:
    An ETI situated inside a black hole might exist nicely without being detectable.

    To state the above as a suitable revised slogan:
    “It is impossible to prove that zero is a real number”

  18. Clic_bron 09 Dec 2008 at 11:01 pm

    “I hate to be pedantic, but this is one of my pet peeves. Logically, it is as easy to prove a negative as a positive (just use modus tonens A–>B, ~B, therefore ~A). ”
    I liked Neil’s response, but I would add that the core assertion that anyone makes regarding the “state of affairs” in the world are making has just two primitive signs. You cannot say that someone is trying to prove N(NA) – that does not correspond to a picture of reality – only NA or A correspond to pictures of reality. Hence it does make sense to say it is impossible to prove NA by any one of the arguments given in this discussion.

  19. Clic_bron 09 Dec 2008 at 11:07 pm

    “I liked Neil’s response, but I would add that the core assertion that anyone makes regarding the “state of affairs” in the world CAN HAVE just two primitive signs. ”

    And on another note, I enjoyed your blog Eric!

  20. Eric Thomsonon 09 Dec 2008 at 11:46 pm

    cwfong said:
    Eric: Can you prove there are no naturally red feathered swans? Or put another way, can you prove I am wrong to assume there are?

    This one example doesn’t justify the general claim that it is impossible to prove a negative, or impossible to prove that something doesn’t exist. I gave examples where this general claim is wrong.

    It is just wrong to say it is impossible to prove a negative. It isn’t a principle of science or logic or reasoning. I look at it as a theistic mantra, which is why I am so surprised it appeared here. It’s like ‘You only use 10% of your brain.’ Somehow it is repeated but it is obviously false.

    As I said in my original post, clearly it is sometimes hard if not impossible to show something doesn’t exist, and the SETI case is likely one of those cases (Niels post focused on the SETI case, not my general point). But there are some in which it is easy (e.g., a ten kilogram lead ball is not in my stomach — I can show you an x-ray to prove it).

    As a general principle it fails. That’s my only point. That’s why I apologized for being a pedant. But please folks, you can prove a negative, or that things don’t exist.

  21. Eric Thomsonon 09 Dec 2008 at 11:56 pm

    “And on another note, I enjoyed your blog Eric!”

    Thanks clic_br.

  22. cwfongon 10 Dec 2008 at 1:32 am

    Well I can agree with Eric that it’s not a perfect axiom.

    Perhaps the phrase should be that it’s impossible to prove something doesn’t exist short of actually proving that if it could, it would.

  23. sonicon 10 Dec 2008 at 7:55 am

    You can prove things mathematically. You can prove things logically. In science we demonstrate things. This is different than proof.
    For example- if Eric produces an x-ray that shows no picture of a cannonball in his stomach, then he has produced evidence that there isn’t a cannonball in his stomach. That is not proof, that is evidence.
    Science doesn’t prove things- it demonstrates likelihood.
    “Nothing can prove my theory correct, but one experiment can prove it wrong.” (that is often atributed to Einstein- it is certainly the correct view of science.)

  24. Steven Novellaon 10 Dec 2008 at 11:27 am

    Eric – you are correct. I was imprecise in my wording. It would have been better to say that it is impossible to demonstrate the non-existence of a phenomenon. This is a practical impossibility, because you would have to survey in adequate detail the entire universe, and by the time you did that the universe will have changed rendering the survey obsolete. Yet, a single example can demonstrate the existence of a phenomenon.

    I will try to be vigilant about avoiding the sloppy shorthand in the future.

  25. Scott D.on 10 Dec 2008 at 3:32 pm

    The biggest problems for SETI seem to be the range of bandwidths for communication and the immensity of space. When, or if, we find extra solar planets with diatomic oxygen that should greatly increase their chances of getting a hit.

    Since astronomers were recently able to detect carbon dioxide in the atmosphere of an extra solar planet we may not be that far away.

    But I’m not holding my breath.

  26. Puppet_Masteron 10 Dec 2008 at 3:33 pm

    So would I be correct in saying: Logically, you can prove a negative. In reality, you can never prove an absolute negative because you are always making at least one assumption. The most basic assumption everyone makes, I believe, is cogito ergo sum (I think, therefore I am).

  27. TheBlackCaton 11 Dec 2008 at 12:09 pm

    Niels’ comparison with elementary particles is a good one. However, one particular example along those lines I think is particularly similar is proton decay. Some proposed replacements for the Standard Model predict that protons will decay into elementary particles with a certain half-life. This is not the prediction of the standard model, but people have been looking very hard to find proton decay events. So far all attempts have failed (there may have been some possible examples that were not reproducible or turned out to be errors). But that does not mean proton decay does not occur, all it can tell us is that the half life must be at least a certain amount (that is it must be greater than that value and less than or equal to infinity). But I would be very surprised (although I suspect Niels can tell us more definitively) that particles physicists do not consider the search for proton decay to pseudoscience.

    However, I have a great deal of difficult seeing any fundamental difference between the search for extraterrestrial intelligence and the search for proton decay. The only difference I can see is that the universally-accepted theories for life and cosmology tells us life on other planets is fundamentally possible while the universally-accepted theory of particle physics tells us proton decay is impossible. That is not a criticism. There is nothing wrong with looking for things the existing theory says are impossible, in fact it is probably the most effective way to proceed when there are indications that the existing theory is flawed (like, in my layman’s understanding, many physicists now think the standard model is). It does, however, make me question why people are so opposed to funding SETI while having no problem with spending massive amounts of money to look for proton decay.

  28. Niels Kjaeron 11 Dec 2008 at 2:24 pm

    I agree with TheBlackCat,

    search for proton decay has always been considered a standard part of High Energy Particle physics. It’s normally been a nice “bread and butter” part of the big underground experiments, which were typically designed (and funded) to look for more “mainstream” effects like neutrino oscillations.

    The Standard Model has always been considered “flawed” in the sense that it is incomplete and must be altered at least at the Planck scale. It can be considered like Newtonian physics which has to be altered at the atomic scale by quantum physics.

    The big experiments that will run at the Large Hadron Collider at CERN will each do order of 50 mainstream searches for “new physics” each considered worth publishing in refereed jounals even when nothing new will been found.

    Because so many physicists look in detail on so much new data, there will also be room for evidence for some pseudoscientific discoveries. But they will go away when more statistics become available. This has always been a natural part of experimental high energy physics.

  29. mindmeon 12 Dec 2008 at 10:50 am

    I’m never sure why people get confused about SETI. The name explains it. “Search”. Any hypothesis starts with a search for an effect or the existence of something not seen but hypothesized. Once you claim to have found something, then falsification comes into play.

  30. […] is analogous to the search for Intelligent Design (ID) in nature. This time he is responding to a recent blog entry of mine on SETI. He doesn’t actually respond to any of my points – he is just using my entry as an excuse to […]

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.