Today’s New York Times science section contains an interview with physicist Sean Carroll, conducted when he went to New York to discuss his new book, From Eternity to Here, on Colbert. I haven’t yet read it, but I certainly will, for I really want to know why time goes in only one direction. Carroll hints at a solution in his interview, but like all good authors he doesn’t give away that solution, which appears to have something to do with entropy:
Q. YOU WRITE THAT THE NATURE OF TIME IS SUCH THAT WE CAN’T GO BACKWARD.
A. It’s likely that we can’t do time travel. But we don’t know for sure. The arrow of time comes from the increase of entropy, meaning that the universe started out organized and gets messier as time goes on. Every way in which the past is different from the future can ultimately be traced to entropy. The fact that I remember the past and not the future can be traced to the fact that the past has lower entropy. I think I can make choices that affect the future, but that I can’t make choices that affect the past is also because of entropy
Carroll is a smart and amiable guy, and gives a good interview. There’s one place, however, where I think he misses the mark. That’s where he discusses the effect of time’s directionality on biology, specifically ageing:
Q. THE CENTERPIECE OF THE RECENT MOVIE “BENJAMIN BUTTON” AND THE ABC TELEVISION SERIES “FLASH FORWARD” IS THE TIME TRAVEL. HOW DO YOU RATE THE SCIENCE OF THOSE ENTERTAINMENTS?
A. Well, the Benjamin Button character ages in reverse. In “Fast Forward” people glimpse the future. These are great story-telling devices.
But the writers can’t resist the temptation to bend the rules. If time travel were possible, you still wouldn’t be able to change the past — it’s already happened! Benjamin Button, he’s born old and his body grows younger. That can’t be true because being younger is a very specific state of high organization. A body accumulates various failures and signs of age because of the arrow of time.
But I don’t think that entropy (at least in bodies) is the only solution here, or even an important solution, for it’s perfectly possible for a body to be immortal, and some plants (and bdelloid rotifers, who appear to reproduce largely asexually) have approached physical immortality. There’s far more to ageing than just “the arrow of time.” Indeed, the inexorable increase in entropy encapsulated in the second law of thermodynamics, a law that holds over the whole universe, is violated locally by two biological phenomena: development and evolution.
There are several theories of ageing. The evolutionary “pleiotropy” theory says that it pays organisms to reproduce earlier rather than later, so genes that enhance early reproduction even if they cause later problems such as tissue senescence will often be favored. There’s also a “physical breakdown” theory: the thousand natural shocks that flesh (or stem) is heir to will eventually wear out an organism so that it simply ceases to be. A variant of this is the mutation-accumulation theory, in which mutations simply accumulate in the somatic (i.e., non-reproductive) tissue over time, and bad mutations that have their effects later in life will be less disadvantagous than those whose effects show up in youth. This could lead to the accumulation of “ageing genes” and hence produce senescence—the physical breakdown of individuals as they age.
There are other theories, too, and they’re nicely summarized in the Wikipedia article on senescence. But I think none of these are solely explained by “the arrow of time” and entropy. If organisms could simply take energy from the environment (ultimately, of course, derived from the Sun’s increasing entropy), it’s possible to repair mutations (this is already done to some extent) or fix bodily damage and prevent ageing. Another way is to reproduce by splitting or by nonsexual reproduction (parthenogenesis), which is practiced by many organisms like the rotifers. Indeed, the fact that an ageing organism can reproduce at all and produce new, non-senescent offspring is evidence against Carroll’s assertion. Reproduction, sexual or otherwise, shows that it’s not entropy alone that causes ageing, for reproduction completely nullifies the ageing process, and, when an old decrepit soul like me produces a child, the increase in entropy is reversed.
Now development and reproduction, as well as evolution itself, do involve a local decrease in entropy. But this doesn’t violate the second law of thermodynamics (i.e., the law that says “entropy increases”) because that law makes a statement about the entire universe, and in that sense is still true when organisms develop or evolve. That’s because, as we all know, biological decreases in entropy that occur when a fertilized egg develops into a more organized adult, or when a replicating macromolecule evolves into a metazoan—come at the expense of an increased entropy of the Sun, whose energy fuels both evolution and development. And as the sun provides energy, it becomes less ordered. Over the whole universe, there is less energy available to do work, and, in the universe as a whole, which is where the second law is meant to apply, entropy increases.
It would be perfectly possible for organisms to evolve self-repair mechanisms that would render them immortal, and many species have gone partway to this end. Salamanders, for example, have evolved the ability to regenerate limbs, and I’ve already mentioned that organisms have evolved complex ways to repair mutations. No matter what theory of ageing you have, it could be reversed without violating the second law of thermodynamics.
This is a roundabout way of saying that ageing is more complicated than Sean lets on. Yes, time’s arrow may be involved in any explanation for ageing (after all, mutations accumulate in one direction over time, and, Benjamin Button nonwithstanding, senescence occurs in only one direction), but I don’t think that senescence is properly explained by saying “being younger is a very specific state of high organization.” Older organisms can produce younger ones, and that itself leads to higher organization over time.
I’m a fan of Sean’s, so this is merely a mild corrective to what I see as a misleading statement about development and, by proxy, about evolution.
Another example is the “immortal” jellyfish!
http://www.telegraph.co.uk/earth/wildlife/4357829/Immortal-jellyfish-swarming-across-the-world.html
Gravity’s cumulative effect on the cellular matrix over time is reason alone to invest an IRA Roth account.
Jerry Coyne taking on Sean Carroll is always a feast — no matter which particular Sean Carroll.
It is interesting to see that the physicist himself seems to become captured by the colloquial association of entropy with order, which sometimes is helpful and sometimes isn’t. And as Jerry Coyne points out, it is all about where you draw the border, which in open living systems is inherently tricky.
How entangled these concepts are is perhaps illustrated by a simple example from biology: At first sight everyone would agree that the correctly folded 3D structure of a protein is the more ordered state than its denatured, unfolded form. Yet, protein folding (as long as it is not assisted by energy consuming enzymes) is driven by entropy. This is because the seemingly unordered, unfolded protein forces a greater order on the surrounding solvent, where hydrophobic side chains impose local structure on the water molecules. When these side chains are tucked away in the centre of the properly folded protein, the water molecules have more freedom to be unordered and the entropy of the entire system has increased.
Very nicely summarized!
Doesn’t your criticism cause problems for the whole idea that our experience of time is due to entropy? Sean says, “The fact that I remember the past and not the future can be traced to the fact that the past has lower entropy.” But the creation of memories is (locally) a loss in entropy. Why would the increase in entropy in the universe as a whole have anything to do with our experience of time here and now?
I would venture the guess that this is because entropy is operant on anything that is not a subatomic particle. (i.e., anything made of huge numbers of particles experiences entropic effects, and consequently, time).
Carroll’s views on time are well publicized. Brian Greene discusses the entropy view in Fabric of the Cosmos. The idea is that the arrow of time is an illusion created by probability. If a container (like the universe) increases in size, then the distribution of particles inside will get more disordered as a matter of probability (because the number of equivalent configurations of the particles increases, so the probability that that they are in any one particular configuration decreases).
So in an expanding universe, even though our equations for describing particles are time-independent, what we actually see follows from what’s probable at any given size of the universe, which follows a trend with a particular direction.
But I agree with you that Carroll is wrong in claiming that reverse senescence “can’t” be true. Locally, it’s possible in principle, just as we can treat other breakdowns of the body with surgery and such.
You’re oversimplifying Greene a bit, he doesn’t say the resolution to the arrow of time problem is solely a matter of the expansion of space. If that were true, then in a universe with sufficient mass to eventually halt the expansion and contract into a Big Crunch, the arrow of time would reverse, but this is not what any physicist would expect (it would require a very special choice of smooth, low-entropy boundary condition at the Big Crunch–as Greene points out on p. 173, when gravity is involved a high-entropy state would actually be very clumpy, so what physicists would actually expect in this scenario is a ‘clumpy’ Big Crunch filled with black holes that would not look anything like the smooth distribution of matter after the Big Bang). Greene’s argument in chapter 11 is a more subtle one, that the quantum process called inflation could take a very tiny region of space with relatively low entropy (produced by a random fluctuation in a larger high-entropy universe, perhaps) and blow it up to a huge smooth region with an enormously low entropy, and then the arrow of time in the universe after the inflationary period would be a process of entropy increase from this enormously low initial state at the end of the inflationary period. An equally huge region of space that had *not* been produced by inflation and was *not* in a low-entropy state would not lead to a universe with an arrow of time, even if it was expanding just like our universe. Also, if you read Carroll’s book “From Eternity to Here” you’ll see he actually agrees with the idea that inflation can give a large smooth initial state and is thus a key component of the answer to where the arrow of time comes from.
Jerry, I agree with most of what you say, although I think you might be expecting a lot out of a single extemporaneous sentence in an NYT interview. Obviously aging is not *just* a matter of entropy increasing. But two important points:
First, aging certainly doesn’t violate the Second Law of Thermo, because nothing in our universe (involving more than a handful of particles) violates the Second Law. The law says that entropy never decreases in closed systems. In open systems, it obviously does; otherwise there would be no such thing as a refrigerator. Living organisms are very far from closed systems, and their lives are a constant give and take between the increase of entropy they would experience alone and the upkeep they manage by taking advantage of free energy from their environments.
Second, in a very grand sense the Second Law is certainly responsible for aging, because the Second Law is responsible for the arrow of time. That is: without the Second Law, we would be in thermal equilibrium; entropy would be maximal and unchanging. But that means *every* macroscopic property would be unchanging. There would be no gradients in space or time — no aging, no evolution, no metabolism, no memory, etc.
No question that my NYT answer could have given the impression that a body was like a closed system that simply decayed with time, which is very far from the truth. Thanks for the elaboration on that. If only newspapers believed in hyperlinking…
Ah, that helps. I was wondering about this as I read Jerry’s comments. I was recalling Atkins’ point in Creation Revisited – I’m a long way from being a scientist, so these ‘picture books’ help – about complexity, and how chaos may temporarily recede… (and here I had to go and look it up), “But these are temporary and local deceits, for deeper in the world the spring inescapably unwinds. Everything is driven by decay.” If it weren’t, as he says, then the universe must be eternal, and there would be no night, every point in the sky would be a star, and the sky would be bright with millions of suns still radiating light endlessly.
Sean, thanks for weighing in, and let me say I sympathize with the plight of having to give a one-sentence answer off the cuff 🙂 However, in terms of your argument that all macroscopic properties are dependent on the 2nd law, I think that is a misdirection on your part. It is trivially true, but so what? As I mentioned in my comment below, if you were asked why objects fall down to the ground instead of up to the sky, would you answer that it was because of the Second Law of Thermodynamics? You could argue that answer would be correct — but it would also be very silly 🙂
James, you say that this is “trivially true,” but it’s not really trivial; it’s only that causality and the arrow of time are so deeply embedded into our way of thinking that it is hard for most people to conceive that it could be different. Actually, the laws of physics are all invariant under time reversal* at the microscopic level. In your example, it really is possible for objects to “fall” up to the sky. The reason for this is profound, not silly.
Suppose you drop a book on the floor. It falls for a fraction of a second and then stops. But why does it stop? Doesn’t this sudden loss of kinetic energy violate the law of energy conservation? Actually, no, because the kinetic energy of the book is dissipated as sound waves, which cause a slight heating of the air, the floor, and the book.
Now in principle it is possible for a book sitting on the floor to suddenly fly up into the air and be launched into outer space. This is perfectly consistent with all of the microscopic laws of physics. All it would take is for the random thermal motions of the molecules in the floor to suddenly “conspire” and all go in the same direction (i.e., up) at the same time. But of course this never happens, and it seems ridiculous to even think about. The reason why it never happens is that, although such a “conspiracy” is indeed possible, it is overwhelmingly improbable. The number of ways for the molecules to move in a manner that we would perceive as “random” is vastly greater than the number of ways that they could all move together.
So before you accuse Sean of making trivial statements, it may be wise to spend some time studying the issues at a slightly deeper level.
*For some interactions, time reversal must be accompanied by other transformations (such as charge conjugation and parity).
I think you misunderstood what I meant by “trivially true”.
In any case, if you are seriously arguing that “Thermodynamics!” is the best answer to the question “Why do dropped books fall to the floor?”, then I don’t know what to tell you.
“Thermodynamics” is not the reason dropped books fall to the floor, gravity and inertia explain this. However, gravity and inertia can also be used to explain the trajectory of an object moving *away* from the floor after it’s been given a kick by some non-gravitational force, like a rubber ball that fell to the floor and then bounced back upwards due to the collision. If the ball’s collision with the floor involved no change in entropy, the collision would be perfectly elastic and the ball would just keep bouncing to exactly the same height, so that a reversed movie of the bouncing ball would look no different from the same movie played normally. Entropy is needed to explain why things tend to come to a state of rest after hitting the ground, and why we don’t see things bouncing up away from the ground just as often as we see things falling down towards it.
In fact, *any* time-asymmetry we see on the macroscopic scale has an explanation involving entropy–try thinking of a single example of a phenomenon that would look “wrong” if the movie were played backwards, but where the explanation for the “wrongness” of the backwards movie does not ultimately reduce to the fact that the backwards movie shows entropy spontaneously decreasing. You can’t do it! According to the best current understanding of physics, any macroscopic “arrow of time” is a consequence of thermodynamics.
I’d have thought that the Times was respectable enough of a paper that they’d be able to interview people without HAVING TO RESORT TO SHOUTING VIA ALL CAPS.
I think his use of entropy as explanatory for the arrow of time is just plain wrong.
The arrow of time is an artificial concept derived from mathematical modelling, not reality. Time is nothing more than a counting of events, and the ability to use negative time values in a formula does not mean that uncounting events is possible, or even meaningful to contemplate.
As Jerry points out, entropy decreases all the time here on Earth, because there’s a huge influx of energy from the Sun. If the “arrow of time” were dictated by an increase in entropy, then we would expect time to move backwards wherever entropy actually decreases.
The very notion of “remembering the future” is just silly, and isn’t even coherent unless you discount the way brains actually work.
But then, whenever you stretch a mathematical model too far in an attempt to access more insight about reality than was put into the construction of the model in the first place, you’re bound to come up with a lot of silly ideas.
I think the argument, though, is that the way brains actually work is constrained by the types of chemical reactions that are possible, and that in turn is constrained by thermodynamics.
I’m not at all convinced that “entropy” is a sufficient, or even a partial, answer to the arrow of time problem. But I don’t think it’s a silly answer at all; it has some real potential.
`I don’t understand you,’ said Alice. `It’s dreadfully confusing!’
`That’s the effect of living backwards,’ the Queen said kindly: `it always makes one a little giddy at first –‘
`Living backwards!’ Alice repeated in great astonishment. `I never heard of such a thing!’
`– but there’s one great advantage in it, that one’s memory works both ways.’
`I’m sure mine only works one way,’ Alice remarked. `I can’t remember things before they happen.’
`It’s a poor sort of memory that only works backwards,’ the Queen remarked.
Thanny, it may be a good idea for you to read Sean’s book before you start calling him “silly” and “just plain wrong.” I haven’t read his book yet, but I have studied others on the same subject, including H. D. Zeh’s The Physical Basis of The Direction of Time (which is probably more technical than Sean’s book). Sean isn’t making any silly mistakes here.
Leonard Hayflick has an editorial in PLoS Genet, Hayflick L (2007) Entropy Explains Aging, Genetic Determinism Explains Longevity, and Undefined Terminology Explains Misunderstanding Both. PLoS Genet 3(12): e220. doi:10.1371/journal.pgen.0030220. This review may help lasering some of the equivocations regarding aging and thermodynamics. A stern warning ends the abstract:” As a consequence of the terminology dilemma, I will define for use in this editorial the four aspects of the finitude of life: aging, the determinants of longevity, age-associated diseases, and death. I will not discuss the latter, although even this word defies a universally accepted definition? It may help some.
Sure, a system can locally decrease its entropy, but always at the expense of increasing the entropy of its direct surroundings. But there is no way that a system can create and maintain perfect dividing line between the region where entropy decreases and where it increases. You can keep back entropy for a long time, but you can’t keep it away forever. No process is 100% efficient.
But in regards to aging, this argument would only actually come into play with some hypothetical organism that outlived the useful life of the sun…! Entropy may explain why true immortality is impossible, but it has very little to do with human aging…
Except that we don’t get our energy directly from the sun. The sun only powers the very first stages of all the chemical processes of life. All the other energy and matter exchanges are completely local, and will influence the local entropy.
So no, this argument doesn’t apply only when the sun runs out. We exchange matter and energy with our environment all the time, and there is no way to be sure that all these interactions will always be to our advantage. There is no way a living organism can keep all the entropy out, as it were.
Why shouldn’t a living organism be able to keep its entropy low indefinitely as long as it was getting food and air and other necessary low-entropy inputs from the environment? The biosphere as a whole does this, as do collections of reproductive cells (any new young organism is just made out of a collection of germ-line cells which spawned off of its parent after all). And some organisms do appear to have no natural aging, dying only due to ‘accidents’ like being eaten (see http://en.wikipedia.org/wiki/Immortality#Biologically_immortal_species for some examples)
I agree that “entropy” is a very very poor answer to “why is a Benjamin Button scenario impossible?” I’d sum it up like this: The arrow of time and the nature of causality are questions of physics, while senescence is a question of biology.
Of course physics underpins biology, and Carroll is correct in the trivial sense that if the arrow of time didn’t work that way, then I suppose any biological phenomenon (or any phenomenon at all) could be reversed. But that doesn’t explain anything about senescence.
It would be like if I asked why objects fall down to the ground instead of falling up to the sky, and you answered that it was because the 2nd law of thermodynamics said that time could only move in one direction. In a sense, that is trivially true: if time were played backwards, gravity would superficially appear to be a repelling force rather than an attracting force. But it explains absolutely nothing, and hence is kind of a silly answer. Much better to explain a falling object using the theory of gravity than by using the laws of thermodynamics!
You need to think about that one a bit more. If you played time in reverse, gravity would still be a force of attraction, not repulsion. If that were not the case then there is a huge fundamental problem with our current understanding of Newtonian mechanics.
I read Sean Carrol’s From Eternity to Here a few months ago and I was not convinced that entropy causes the arrow of time.
I found the book to have a great start and ending but the entropy chapters in the middle were not clear.
Too much uncertainty in the principle?
And remember – time flies like an arrow, while fruit flies like a banana.
That’s funny!
Of course it’s funny! It’s from Groucho Marx.
I don’t think that’s possible.
First, I don’t think it’s possible to have perfect self-repair mechanisms to begin with. What if the repair mechanism itself breaks? Do you then develop self-repair-mechanism repair mechanisms? But what if those break down too?
Second, even if perfect self-repair mechanisms could exist, I doubt that they could evolve. Experience with designing fail-safe systems shows that the incremental costs skyrocket the closer you approach 100%. At some point the fractional benefit just can’t justify the huge increase in costs. Evolution seems more in the business of producing “good enough” solutions, not perfection. Why would a self-repair system be the only exception to evolve to perfection?
Evolution has no need for perfect self-repair mechanisms. Once an organism reproduces successfully, it (eventually) hinders further success due to competition with others of its species (limited resources).
That’s exactly what I think.
Why are you comparing biological organisms with machines?
Biological organisms are machines. But that is not relevant anyway.
The comparison with machines made by humans is way off though; the analogies just don’t work.
No, it is not.
Radioactive decay renders it impossible.
Eventual(!) proton decay makes it impossible.
Quantum fluctuations make it impossible.
Heat death of the universe makes it impossible.
Universal expansion makes it impossible.
Lawrence Krauss & Freeman Dyson have gone into this is excruciating detail.
I assume that you may have not meant ‘immortal’, but more akin to ‘very, very, very long lived’?
Why would radioactive decay make immortality impossible? There are many abundant elements which we have never observed to decay. As for the sun evolving into a red giant, surely the immortal beings would have escaped into the cold space between stars? So, you’ll have to rule out radioactive decay and the sun as definitive show stoppers.
If time travel were possible, you still wouldn’t be able to change the past — it’s already happened!
But doesn’t that also make time travel impossible? Because the past has already happened, without you in it!
I’m probably missing something really obvious…
Three words: Branching time lines. Sheesh, haven’t you ever watched Back To The Future? 😉
Yeah, I don’t buy that one either. From a philosophical standpoint, how can anyone in the future know that someone didn’t make some alterations to the past? Presumably only those who made the alterations would really know. For all we know, Hitler won the war but Dr. Who altered the outcome of some events and encouraged the development of other events which eventually led to Hitler’s defeat; so the original history was altered without our knowledge.
The idea is that if you time traveled to the past, you were always in the past during that time, and you always had some effect. It never happened without you there. e.g. if I travel back to 1914 to stop the assassination of Archduke Ferdinand, I won’t be able to, because I was there when it happened and I wasn’t able to.
… It’s surprisingly hard to explain even though I feel like I understand it well. Just watch Lost 🙂 It’s the opposite of the Back to the Future idea of time travel: nothing about the timeline can change.
I’d just like to point out that there is not even a ‘local’ violation of thermodynamics in organisms; entropy is increasing. If you do half-ass accounting and don’t figure in a cell’s energy sources then you may make the (incorrect) claim that there is a net decrease in entropy. However, Carroll’s claim that ageing is due to increasing entropy is meaningless. Entropy does not cause anything; it is a result of processes.
Don’t forget the entropy in the waste products too.
The bible sez cells don’t have waste – imagine billions of cells pooping in you all the time! How silly!
Indeed, it’s laughable. Until your kidneys stop working, of course.
Oh, and yes, aging is a process. But the laws of entropy do tell us something about the reversibility of such a process.
The concept of entropy can help us understand why processes are not ideally reversible, but entropy in itself is an effect and not a cause or a reason for anything. Unfortunately Carroll’s choice of words is too easily misunderstood by people who don’t already know the ideas well.
…and by a coincidence, my copy of “Timeless Reality” by Vic Stenger just arrived. And now I have to get’s Sean’s book, too.
So many books, so little time. If only I could figure out how to behave like a subatomic particle.
well…immortal species always were possible….and one would think immortality would be a fitness benefit in the sense it increased reps by extending fertility.
but that decrease in genetic variability causes extreme sensitivity to environmental changes….and the environment is definitely subject to entropy.
so aging is an optimization algorithm in a sense…a trade off between genetic adaptibility and number of reps generated.
the selfish genes code for three things…reproduction, survival, and death.
individual organisms are programmed to die.
but….since homo sap can impose order on our environment now…we can in theory cheat death by a lot….and become virtually immortal.
My professor thought about 200 years– but with organ replacement, cloning, and eventual digital personality reproduction immortality is certainly doable.