One of the things I love so much about the Drake equation is how it's all so much bullshit.
Before you downvote me, I think it's a great equation and I don't have problems with folks using it. As long as you know what you are doing. really, think about it, all it actually says is that there is this list of things that we feel have to happen in order to determine communicating civilizations, and given any random set of probabilities for a list of things, you multiply them together to get an aggregate answer. It doesn't address whether the list is complete, or overly complex, or mis-stated. The numbers provided are juts best guesses. It's just "multiply a bunch of numbers together and get a new number"
I love it. It's definitely not science -- yet it has this cool-looking formula involved. It also has a powerful impact on folks who use it.
Drake is really cool in that it gets people thinking about how likely ET is. For that, it's a wonderful tool. But at its heart there's really not much there. At least yet. Perhaps in the future we'll have some solid survey results to put in, especially with Kepler online. Can't wait to start seeing those results coming in.
I really dislike the drake equation because it's the epitome of bad science -- when someone enters value for the variables they have no idea what is reasonable, absolutely no idea -- is it 0.1 or 0.0000000000000000000000000000001 we have no way of knowing.
Though I do like how it gets people to realize how small and insignificant we are in the grand scale that is the solar system, the reasoning we take to get there is fundamentally flawed. We use numbers that have everyday meaning to us, and of course it spits out nice results; this is flawed reasoning since we have no basis to think those numbers are even remotely correct. In general, I just think it's sloppy thinking that takes advantage of deficiencies in human thinking -- it's a magic trick.
I really tried to be even-handed with my comment -- I didn't want to sound like a troll -- but you are completely correct.
In fact, the Drake Equation is a religious statement. That is, it is a creative speculation using commonly understood symbols to that which are completely unknown to us. It's the same as a couple of cavemen sitting on a hill talking about thunder as being large rocks falling from a cliff in a faraway valley. It makes sense, it fits in with what we know, but it's really just a bunch of guesses connected with symbols that, based on context, we feel social approval towards. It's a mind-hack. And a very nicely executed one at that.
The reason I didn't use the "R" word, religion, is that folks see that word and stop thinking. They give you an instinctual answer. No way! I'm not religious! etc.
The Drake equation basically says what anybody who has looked up at the night sky knows intuitively: there are a lot of stars out there, and it's extremely unlikely that we're all by ourselves. It just does it in a way that makes us feel like we're receiving some sort of imparted scientific wisdom, instead of what it really is: creative speculation. In that sense the formula itself may be total bullshit scientifically, but what it tells us about how science and religion play with each other is very important. People look at it and just feel that there is some truth there, and there is, but it's not the same kind of truth as f=ma, although both statements are expressed mathematically and have the same air of seriousness around them.
"The Drake equation basically says what anybody who has looked up at the night sky knows intuitively: there are a lot of stars out there, and it's extremely unlikely that we're all by ourselves."
Further reinforcing your core point, it can also tell you that we are. You don't have to be that painfully conservative with the numbers to come with one or fewer per galaxy. For instance, 50% of stars having planets may be a brutal overestimate, a lot of stars are in multi-star systems or other situations in which there are no stable orbits, resulting in any initial planets getting ejected very quickly if they form at all. 2 Earth-like planets per star is a bold statement, depending on the exact definition of "Earth-like", we only have one in our system and if you read past the propaganda in the planet finding press releases they aren't coming up with all that many other "Earth like" planets yet. Yes, that's because they still can't see them, but the press releases convey a false sense of how the search is going when they do things like call a planet with a surface gravity of 3 or 4 G and a surface temperature beyond rock's liquification point a "Super-Earth". 10000 years is a bold estimate for how long a civ will be around and communicating, certainly. I actually am against the self-loathing that some circles find fashionable but 10,000 years is a long time. I consider these numbers generous and it still only comes up with 10.
It's missing some terms, too. Irregular galaxies [1], approximately 25% of the galaxies out there, are thought to be entirely unable to support complex life because basically all the stars in such galaxies will take a system through a region of the galaxy with too much radiation and sterilize the planets every few tens of millions of years. I mean this only as an example, you can go on naming more fairly reasonable criteria that eliminate half or more stars from contention for quite a while (star types, star galactic orbits, other galaxy attributes, even before we get to planets we can eliminate a lot).
To brutally abuse notation, the night sky is O(n^3) in size, if the probabilities of developing civilization are O(2^n)-type it doesn't matter how large the sky is. Big numbers fall fast when you multiply independent probabilities together, and it's not sophistication to just give up and declare there must be a lot of other civilizations out there; it's still a mathematical copout. We don't know.
I agree with your poing, but Mars is earth-like and it could have carried life ...
There is evidence that Mars once carried liquid water on its surface, it's formed out of the same materials (carbon, hydrogen, water) and had all the elements needed for life. Mars lost its liquid water because (contrary to earth) it lost its atmosphere because it also lacks active plate tectonics (also, contrary to earth).
So our solar system has 2 earth-like planets ... with one that became a desert and died, but that was once very much like Earth, with some people still hoping to find microscopic life at the poles.
That's not too shabby. Basically it is considered that if a planet holds liquid water, then it can carry life (liquid water implies other things, like the presence of an atmosphere). Bigger solar systems may contain even more earth-like planets, but I agree: an average of 2 is way too high.
EDIT:
Also, one other thing: if Mars wouldn't have lost its atmosphere, then oxygen would have spread faster than on earth, because the planet is smaller, which means the transition from microscopic life to macro would have been faster.
That's what I meant by "it depends on your definition of Earth-like". Mars and Venus are at least debatable, many of the planets being reported as "Earths" in the media aren't. (Yet. I freely acknowledge this will change, and soon. I have no doubt that we will find rocky planets of the approximate right size in the approximate right orbit in the near future. Getting a bead on their chemistry and such is going to be a lot harder for a while, though, but we'll crack that too, eventually.) I tend to favor "actually capable of supporting life" but the equation doesn't seem to actually require that.
Mars has the strike against it that it is arguably on the frozen side of the zone of life. Earth is actually already on the somewhat cold side and has spent a fairly large portion of its life as an ice planet; some have speculated that our frequent ice ages have contributed to our diversity of life by moderately predictable periodic mass extinctions, but of course this is just a theory in all the bad senses of that phrase. Venus is possibly too far on the hot side but I have an easier time envisioning life on a relatively hot planet than a relatively cold one. We have extremophiles here on Earth and who knows what they'd be able to evolve into if they weren't in such a small niche? Whereas cold really puts a stopper on life; a critter can evolve that can survive being frozen and possibly even carry on simple life functions but only at a very slow rate, and for all we talk about how fast and robust life is it does not take very much slowdown of life before you can't get an intelligent civilization evolved before the sun ceases to support life. Life is thought to be ~3.5 billion years old here, and it's hard to pin down exactly how long the sun could sustain non-intelligent life but it could be as little as ~1 billion years. Cold life has a real challenge getting to intelligence in time. (Also why I don't spend any time wondering about life in nebulas; they may be a science-fiction staple but at the rate they could live they don't have enough time to evolve into anything interesting before the heat death of the universe.)
Venus is also very Earthlike (potentially much more than Mars in the past) and could have had liquid water and life before the runaway greenhouse effect.
> It's the same as a couple of cavemen sitting on a hill talking about thunder as being large rocks falling from a cliff in a faraway valley.
Those cavemen are doing rudimentary science. They've come up with a testable theory that explains an observed phenomenon of the natural world. If they were to conduct surveys of the cliffs within hearing distance before and after thunderstorms, they'd eventually falsify the theory.
> It just does it in a way that makes us feel like we're receiving some sort of imparted scientific wisdom
If that is the way people choose to see it, that is not Drake's fault. It sounds like you and dantheman really only have a problem with the bullshit "religious" implications some poor souls wish to divine from a dry bit of math.
The Drake equation is not a religious statement until you fill in faith-based numbers.
My only problem is that it gives people some senses of rigorousness, without actually providing it. I could say the chance of extraterrestrial life is .4% pulling that number out of the air, and you can use the Drake equation and get .99% but you know what -- both are made up #s and yours is in no way more accurate than mine, but because of the use of an equation people will think it has scientific merit and is more rigorous.
It's not bad science, though, because it's not science. It's not meant to be science. All Drake set out to do was establish an agenda for an NAS meeting in 1961 regarding extraterrestrial intelligence. It's a list of known unknowns. It was not arrived at via scientific experimentation, and was never claimed to have, so anyone calling it "science", even "bad science", is mislabeling it.
There is a wonderful book edited by Carl Sagan called Communication With Extraterrestrial Intelligence (CETI), Proceedings of a conference held at the Byurakan Astrophysical Observatory, Yerevan, USSR, 5-11 September 1971.
The chief aims of the conference were, first, to estimate values for terms of the Drake equation, and second, to devise strategies for contact. So all these top American and Soviet scientists had to discuss a remarkable range of topics—everything from the chemical origins of life, to the anthropology of early civilizations, to the problem of encoding messages in a form comprehensible to intelligences vastly unlike ourselves.
It's possibly the most interesting thing I've ever read.
Why not? It's proposing an underlying model, which in turn provides a framework for investgation as well as a way reasoning about things in the absence of harder data as. Sounds like science to me.
What would you propose as a more scientific approach to estimating the number of communicating civilizations in the galaxy?
The Drake equation is a lot like the Fermi paradox. They are not revelations of some truth about the nature of intelligent life in the Universe (or even our galaxy), they are merely reflections of our utter ignorance of the subject.
Where do these figures come from? For fully 5 out of the 7 parameters to the Drake equation we have extraordinarily limited data. How common are Earth-like planets? Call back in about 5-10 years when we'll have some solid data from the Kepler mission. How common is life on Earth-like planets? Who knows? We aren't even close to being able to get solid figures on that. If we can find life on Mars or rule life out of ever having existed there then we'll have a teeny tiny bit of data. But even then such an estimate will be biased by the anthropic principle and at best only accurate to perhaps a single order of magnitude (if that). Even then that leaves half of the parameters as wild ass guesses.
Not to mention that the parameters aren't independent. If we find life on Mars existed or still exists that would make f-sub-l higher, but it would probably make f-sub-i correspondingly lower.
Worse yet, the very premise of the Drake equation is at odds with the premise behind the Fermi paradox. If even a single, slow colonizing intelligent civilization ever existed in our galaxy it would make the number of possible communicative worlds differ by perhaps even 6 or 9 orders of magnitude.
The exercise is only worthwhile when it is not taken seriously, which is a bit of a catch-22.
The Drake Equation has so many fuzzy unknowns that you cant get meaningful information out of it, if rephrased right we could get some meaning variables to play with. If we want to know if there are possible communication partners right now, we can figure out some of the unknowns. First the definition of an earth like planet. It has liquid water and dry land. Liquid water, because the only kind of life we know of developed in liquid water. Dry land because developing a communicating technology requires an abiulity to isolate your experiments. I believe there are so many impediments under water that a technological civilization would never develop. So a technological civilization requires 4B years to develop from the birth of a star. The habitable zone of a star is where water is liquid roughly. An earth-like planet must have inhabited a continuously habitable zone for 4B years. A flare star may not have a continuously habitable zone. An M star's habitable zone may be so close to the start that the planet is in a tidal lock with the star. An enterprising astrophysist can probably estimate the probability of a sufficently large stable habital zone for a given class of stars. We dont have enough
data for estimating the probability of a water/dry land world being there.
Given life and enough time, what is the probability of a technological civilization arising. 1.0 or so small that we are alone. We are probably at the age and interconnectedness that we will survive a very long time. I
dont wory about nuclear war, but I do worry about bio-terror and nut cases.
I think the big constraint is patience. I dont think any civilization is ready to invest the time and energy to wait 1000 years for a reply. I think the potential partners will be limited to a volume of 100 light years.
Maybe 5000 stars, not all that many.
So the solution is to add a variable for confidence in the completeness of the equation. Perhaps range 0.5..1.5. And then another for confidence in the confidence.
The problem is that we don't know what the confidence should be, and as of right now we don't have anyway to tell. In the future hopefully we'll know enough to use the equation but right now we just don't know.
Its interesting that Alpha has the equation baked in. What would be more interesting would be to see a regularly updated version of this with the latest cosmological data coming in from telescopes like Kepler. Over the next 3-5 years, Kepler is going to be finding Earth-size planets that exist in their star's habitable zones.
For those not familiar with how Kepler is finding planets, I'll explain. Basically, Kepler looks for planets transiting in front of their stars. The telescope watches for dips in the star's light. If the planet has an atmosphere, in some cases it can even get an idea of some of the gases from the light.
The reason that astronomers aren't expecting to find Earth-size worlds in the habitable zones is that they have to wait for several transits to know for sure they've seen a planet. The reason that most of the planets found to date have been gas giants orbiting very close to their stars is that it only takes a few weeks or months for them to transit the star several times. Astronomers can see one transit, but have to wait for a 2nd to hypothesize a planet is found, and have to see a 3rd or more transit to verify the planet is real.
So Kepler has likely already detected suspected Earth-size planets in habitable zones around at least a few stars. However, the telescope has to watch for several years to confirm the transits. Imagine an astronomer in another solar system that has seen Earth transit the Sun. They would have to wait 3-4 years to see Earth transit the Sun several times to confirm our existence.
This is something I have hard time understanding. Most of the stars are having star flares most of the time. If they're like our Sol, they have periods of high and low flares activity, but within the periods the pattern is random (i.e. noise). Presumably, the change in luminosity caused by a random flare is bigger than by a planet transit (remember transit of Venus?). So how are they filtering the noise out?
There's no quick answer to that question, unfortunately, because if you want something other than handwaving the answer is to go learn about the basics of signal processing. But in short, the signature of a set of random flares and a regular crossing of a planet are sufficiently different that they are easily separated by the correct tools, the frequencies are totally different.
That's a really good question.
From my understanding, solar flares themselves can't be detected optically yet. Note: It may very well be possible that Kepler can do this, but I don't know. I'm just an amateur.
In one instance of extrasolar flare detection, http://www.nasa.gov/mission_pages/swift/bursts/monster_flare..., it was found by SWIFT detecting high x-rays coming from II Pegasi.
Basically, this is why detecting multiple transits is necessary. Its entirely possible to see some stellar activity that looks like a transit. However, its very unlikely that multiple observations of transits that match each other can be attributed to solar flares.
What's really depressing is if you assume the number of communicating civilizations is 1 and make average lifetime of communicating civilizations be the free variable.
note that the drake equation gives the number of currently communicating civilizations.
if you think there are 10 communicating civilizations now, each communicating for 10000 years, then there should have been quite a few such civilizations in the 13000000000 year history of our milky way galaxy.
if the parameters provided apply for 1% of the milky ways history, we would expect 13000 communicating civilizations in total.
i would imagine any communicating civilization lasting 10000 years would also develop self reproducing space probes, thus becoming a permanent feature.
When first stars appeared they were made out of hydrogen. It seems that you have to wait for second or third generation stars (formed from rests of previous defunct stars) in order to have heavy elements that would form planets and allow life.
We could be the first, or one of the first civilizations in the galaxy. Trouble with the equation is that it does make nobody happy. As we are knowing better some factors (like how many stars have planets) it seems that life should be everywhere.
But it's enough that one the still unknown unknowns is worse than we think, to make the density as low as a few dozen civilizations in the galaxy. That's too far for us to communicate in a 10.000 years scale.
BTW, I find this 10.000 number is cruft from cold war era. Some people love the extintion idea, not sure why.
The best response to the Drake opinion, IMO, is Tipler's point that if even one civilization created and sent out a von Neumann machine just a million years ago, they'd be here by now. So the fact that we don't see them all over suggests that they do not exist at all.
We are? I find that a rather bold assertion. The main problem of the von Neumann machine is gathering and transforming enough materials on its own. If we need enormous amounts of ore and factories to do that, so does the machine. It would need to be mindboggingly huge, also because it needs enough fuel to reach new sources. A tiny machine won't work, because it can't refine it's own materials. Really, the practical problems are being hugely underestimated.
The Drake Equation's most telling fault, for me, is the "average lifetime of communicating civilizations".
The default is 10,000 years. That results in 10 civilizations.
A 1,000 year lifespan means there is only 1 civilization, and a 100,000 year lifespan means there are 100 civilizations.
As if we have any idea how long a communicating alien civilization would last. We don't even know how long we will last, and aliens would be completely different to us.
A quick thing to notice, if you fiddle a bit with the variables, you can get the answer "1" to the number of civilisations in our galaxy. Just decrease the probability of life or intelligent life. After all, it took 3 bn. years before the first multicellular life developed on Earth and 1bn. years for intelligence to develop afterwards.
Why the search of ET Intelligence is of any significance for mankind (or even science for that matter)? Even if we assume that ET life exists, in all likelyhood, after spending billions of trillions of dollars, we may get to know that ET exists (assuming they do), are thousands of light-years away from us.
Apart from the curious "are we alone?", does answering this question, has any positive/productive outcome?
We are nowhere near billions of trillions of dollars spent on SETI (I'd venture to guess it's in 100s of millions for all these decadesat best), but.
Finding an ET would be event of scientific significance that can hardly be compared to anything else. Perhaps a manifestation of God's existence comes closest.
These are the values originally used by Drake et. al. when they proposed the equation in 1961. Many were outright guesses. Nevertheless, 50 years later, we still don't have much better guesses, but this may change soon for proportion of planets with stars (is probably higher) and stars with earthlikes (coming soon).
When the equation was first proposed, I'd guess they assumed Sol is an "average" solar system, and given that we have Earth, which is by definition Earth-like, and Mars, Venus, and several of the Jovian moons, which could all fit a suitably loose definition of "earth-like", Drake might even have thought he was being conservative.
More interesting to me is the chance of intelligent life variable. A couple of months ago on Ars Technica iirc, there was published a writeup on a paper on how macro life (much less intelligent life) requires something analogous to mitochondria for the cell to be large enough to handle that level of complexity (the problem being caused by the geometric relationship between volume and surface area). the paper goes on to say that the chances of one organism entering another and them then entering a mutually beneficial symbiotic relationship as being astronomically low (it is, however, a function of time, but Ars implied (iirc) that this doesn't noticably help the Drake equation the chance is that low). Based on this I would adjust Fi by a couple dozen orders of magnitude
On a scale broader than single cells there are around 80% of vascular plants and their symbiosis with fungi, the bacteria in our guts, the worms close to black smokers, certain sea cucumbers that absorb algae and live off their photosynthesis, and much more.
I'm okay with the Field of Dreams interpretation of astrobiology: "If you build it, they will come", so to speak. All you need is for life to take hold once and it'll take over a planet in just a few billion years. Of course, my own hunch is that suitable chunks of rock (planets, moons, etc) are somewhat more rare than the default value[1].
[1] I plugged in .1, among other variable changes. On the other hand, I got the number of communicating civilizations = .02, so either I'm too pessimistic or we're just really, really lucky.
I think there are far too many happen chances in evolutionary biology to say that an advanced civilization is inevitable; my rational mind has a LOT of trouble accepting that our current development is a long history of happen-stances beginning from life´s first development, but I see no viable alternative. But to think that given another Earth with life it would happen again? With all of the stretches that happen all along the evolutionary tree I find this absurd.
I concur. Fortunately, the Drake equation gives us separate variables for "develops life" and "develops intelligent life". The value I used for f_i was somewhat lower than they gave, but I also thought that f_c was higher, so it worked out as a wash.
Actually the only thing we know for sure is that life on Earth happened at least once. It is entirely conceivable that several alternatives to RNA-based life (precursor to DNA-based life) popped up in several different places but then were hopelessly out-competed by the RNA-based lifeforms. I am a biologist by training, and the more I studied biology and evolution, the more life appeared to me as a powerful (somewhat scary even) and inevitable process, rather than a "unique snowflake" that many of us commonly think it to be.
Life, given right conditions and enough time, seems to be about as inevitable as the eventual appearance of viral cat pictures on online forums, in other words.
I think there is little wrong with assuming that the probability of emergence of life given right conditions and enough time is one; however we need to be very careful with estimating how common (and how stable!) those conditions are in our galaxy in the first place. A good place to start would be to look at how many places similar to Earth are out there (which we are already doing).
I think there is little wrong with assuming that the probability of appearance of life given right condition and time to develop is one
Fair enough. It's a bit like the infinite monkeys in infinite typewriters scenario. Given enough time, sure, you'll have life. It's a question of how much time. Still, assuming all Eart-like planets will actually develop life given the time lapse of the existence of the Milky Way does still seems like a bold assumption.
I am not a geologist, but the Wikipedia article on the Late Heavy Bombardment (http://en.wikipedia.org/wiki/Late_Heavy_Bombardment#Geologic...) says that 100 million years is the time that an Earth-sized body would take to to cool down and form a crust (the bombardment basically tuned the Earth into a ball of molten magma).
I assume "Earth like" in this context means planets that resemble the only planet known to hold life at least enough to be capable of holding life themselves. The problem is we don't even know what that means exactly. It's also assuming that life can only exist in "Earth like" conditions.
Also, if Fraction of Earth‐like planets that develop life = 1 and Mars is an "Earth like" planet, are they assuming life evolved in Mars too or something?
in that case, value of 1 for "fraction of Earth‐like planets that develop life" is odd since we are far from confident that life ever developed on mars.
What we call atmosphere in Mars would pass for a very good vacuum in a classroom. Life could develop on a Mars sized body, but then the star would have to be far less active than our sun (so that the planet could retain a reasonable atmosphere)
Before you downvote me, I think it's a great equation and I don't have problems with folks using it. As long as you know what you are doing. really, think about it, all it actually says is that there is this list of things that we feel have to happen in order to determine communicating civilizations, and given any random set of probabilities for a list of things, you multiply them together to get an aggregate answer. It doesn't address whether the list is complete, or overly complex, or mis-stated. The numbers provided are juts best guesses. It's just "multiply a bunch of numbers together and get a new number"
I love it. It's definitely not science -- yet it has this cool-looking formula involved. It also has a powerful impact on folks who use it.
Drake is really cool in that it gets people thinking about how likely ET is. For that, it's a wonderful tool. But at its heart there's really not much there. At least yet. Perhaps in the future we'll have some solid survey results to put in, especially with Kepler online. Can't wait to start seeing those results coming in.