Voyager 1, now 17 light hours from Earth, continues to be my touchstone when asked about getting to Alpha Centauri — and in the last few days, I’ve been asked that question a lot. At 17.1 kilometers per second, Voyager 1 would need 74,000 years to reach the blistering orb we now believe to be orbiting Centauri B. Voyager 1 is not the fastest thing we’ve ever launched — New Horizons at one point in its mission was moving with greater velocity, though no longer, and the Helios II Solar probe, no longer functional, reaches about 70 kilometers per second at perihelion. But Voyager 1 will be our first craft to reach interstellar space, and it continues to be a measure of how frustratingly far even the nearest stars happen to be.
Cautionary notes are needed when a sudden burst of enthusiasm comes to these subjects, as it seems to have done with the discovery of Centauri B b. What we need to avoid, if we’ve got our eyes on long-term prospects and a sustained effort that may take centuries to succeed, is minimizing the challenges of an interstellar journey. Making it sound like a simple extension of existing interplanetary missions would create a public backlash once the real issues become clear. Better to be straightforward, to note the vast energy budget needed by an interstellar mission, the conundrum of propulsion, the breathtaking scope of the distances involved.
Image: The night sky above Mt. John Observatory in New Zealand, where a second active hunt for Alpha Centauri planets has taken place. This and a third Centauri project, Debra Fischer’s work at the Cerro Tololo Inter-American Observatory (Chile), give us hope that Centauri B b may be confirmed in the near future. Note the Southern Cross (just above and to the right of the dome), with Alpha and Beta Centauri the two stars to its left. Alpha Centauri (the leftmost bright star) is a triple system made up of Centauri A, Centauri B and Proxima Centauri. Credit: Fraser Gunn.
Marc Millis told Alan Boyle just after the Centauri B b discovery that manned missions were almost certainly not going to be our first attempts to reach the stars. We can work out scenarios where the global economy grows to the point where a civilization moving into the Solar System can afford to build huge laser installations that could propel a lightsail to perhaps ten percent of the speed of light. Get up to that kind of velocity and you make the Centauri crossing in less than half a century, but only for a flyby. Deceleration is another matter, and even if, as Robert Forward showed, there may be ways to decelerate a lightsail at destination, the mission then extends to at least a century and, given our assumptions, perhaps a good deal more.
So we’re likely talking about robotic probes which, given the development time needed to create and launch them, would probably be ‘manned’ by an extremely sophisticated artificial intelligence. In the same article, astrobiologist Dirk Schulze-Makuch (Washington State) told Boyle that an interstellar mission could be seen as part of a roadmap that developed incrementally:
“The roadmap that we have takes a grand perspective, with the objective to scout out our own solar system first, put a permanent human presence on Mars, look at asteroids, and really work first on our own solar system before we take the next step to an extrasolar planet.”
And as I’ve speculated here on more than one occasion, a developmental model like this may take a more gentle trajectory than we’ve previously assumed. A civilization that masters living in nearby space by developing large habitats that can house thousands may eventually branch off from its Earthbound human roots to thrive in the places between planets and stars. Generations may indeed live and die in such habitats not because of a specific mission constraint but as a matter of choice. And if this occurs, a mission to a nearby star may be an organic extension of space living, its destination motivated by long-term curiosity rather than any plans of settlement.
Centauri B b is hardly a compelling target for any kind of mission, but the enthusiasm it has already generated points to the latent exploratory impulse that may be triggered if we do find a habitable world somewhere among the Centauri stars. In another recent piece on the discovery, Ian O’Neill quotes Robert Freeland, deputy project leader for Project Icarus, the ongoing re-design of the 1970s Project Daedalus starship study. Like me, Freeland is pondering how the public would react to the next round of Centauri discoveries, assuming such are made:
“I have often imagined the day when scientists directly image an Earth-like extra-solar planet. We would be able to determine the planet’s atmosphere and surface temperature from its spectrum, and we would thus know whether it might be able to sustain life as we know it. I suspect that once such a discovery hits the news, people worldwide are going to demand that we send a probe to determine whether the planet has life (of any type) and/or could be suitable for human habitation. If the latter proves true, then a manned mission would eventually follow.”
A lot can happen, of course, between the surging interest such a discovery would create and the realization of how tough a mission like this would be. But Freeland is surely right that Centauri B b is a warning shot that tells us we’ll be finding much more about this system in the not so distant future. While it is true that the HARPS spectrograph that made the recent discovery possible is capable of detecting planets no smaller than four Earth masses in the habitable zone of Centauri B, Stéphane Udry (Observatoire de Genève) said in the recent news conference that ESO was already working on technologies that would extend that range down to one Earth mass.
Image: An artist’s impression of Centauri B b. Credit: Adrian Mann.
Meanwhile, supported by the Planetary Society, Debra Fischer (Yale University) has been working on a separate search for Alpha Centauri planets, one she recently discussed with Bruce Betts. Fischer notes how faint the 0.5 m/s signal seen by the Geneva team is and points to the need to confirm the discovery. Her team, which has been talking with the HARPS researchers, has been analyzing its own dataset and running simulations to test detectability. Says Fischer:
“Our best data set for aCenB begins in June 2012, when we completed some stability upgrades to the new spectrograph (CHIRON) that we built for the 1.5-m CTIO telescope (with NSF MRI funding). Our precision since the upgrade matches the HARPS precision, but yields a 5-month string of data compared to the 5-year time baseline of data from HARPS.”
And she adds:
“We are in an excellent position to follow-up, but that will likely require an intensive search over the prospective orbital period of 3.24d when the star rises again in January 2013.”
Would the discovery of a habitable zone planet around Centauri B, perhaps occurring some time in the next ten years, have the effect Freeland believes it would? Given the state of our technology, we couldn’t get off an interstellar mission any time soon, but I would like to think that such a finding would be a spur to research that might turn interstellar studies from a back-burner activity into a more visible and better funded effort. We need to advance the technologies that will help us with in-system projects, from long-term life support to advanced robotics. And we can hope that eventually, a Solar System-wide civilization may emerge that will have the strategies and the energy budget to consider a close-up look at such a distant, tantalizing target.
I strongly agree with your statement:
“Making it sound like a simple extension of existing interplanetary missions would create a public backlash once the real issues become clear. Better to be straightforward, to note the vast energy budget needed by an interstellar mission, the conundrum of propulsion, the breathtaking scope of the distances involved.”
Nice article. I’m glad I found your blog.
Regarding getting there and interstellar travel in general. The power and fuel requirements are of course a function of the mass of the probe… say a light sail or other system. The mass must be extremely small to facilitate acceleration to meaningful fractions of light speed but at some point deceleration will be necessary. Given these restrictions one way around having to send a device without the mass of the exploratory sensors and other devices needed for mission success (a la Curiosity) is to restrict the package mass to only softwareAI/computation hardware, a simple sensor array to facilitate landing, a landing shroud and a nano-bot assembler seed stock.
Most deceleration will have to come from gravitational effects but the mass will be so low that some late deceleration will be through atmospheric friction as is the case with some very small (1 gm) meteorites. On arrival the protective shroud opens and software and assemblers use material from the surface to construct the devices needed to explore the planet. In this manner large exploratory robots and even flying vehicles could be mounted as they “grow” out of the surface materials. It might require a hyper accelerated evolutionary process that leads to more advanced material combinations up to the necessary tolerances. Of course we are no where near this tech yet but may be in a hundred years. How small could the probe be… a few grams (apart from the sail or propulsion unit)?
I would be interested in discussing specifics if you find this idea potentially useful.
ESA has just selected a exoplanet mission Cheops to fly in 2017
http://www.esa.int/esaCP/SEMXFG4S18H_index_0.html
Andrew writes:
This is quite similar to ideas Robert Freitas has published about using nanotechnology delivered through tiny ‘needle’ probes to essentially build an observing station in the destination system. Freitas has also examined some of the ethical questions involved in terms of transporting assembler technology like this into a setting that may include indigenous life forms. Fascinating stuff. The beauty of building in situ is that nanotech would allow extremely light payloads, helping hugely with the propulsion problem.
Fascinating blog and reading experience; thanks Paul.
I am a lawyer with no background in science, but I try to follow as much as possible and to understand latest developments (so I hope my comment would not be silly:)). As discussed many times in CD interstellar travel imposes myriad of challenges to us. One (small, but important) challenge is tools of communication and the speed of light barrier. I mean, even if the signal travels by c it will take at least 4 years for it to arrive here if our crew (or probe) has reached Alpha Centauri system.
I am touching this issue because of exciting discovery of Alpha Centauri B b (which leads us to dream of reaching it some day) and recent Nobel prize in physics for measuring and manipulation of quantum systems. If I understand it correctly physicists have found out the way how to measure entangled particles without disturbing them.
Now if I have understood correctly until now scientists believed that quantum entanglement could not be used for transmitting the information – for the observer would not know in what state the observable particle is before observing it (therefore no possibilities to transfer particular parameter on second particle which is, for example on space ship en route to Alpha Centauri).
Maybe I am wrong, but – could the Nobel prized discovery help us to develop FTL communication systems?
These thoughts were of course provoked by fascinating find of Alpha Cen B b and dreams of getting there someday.
I now I could be wrong but would be glad for any explanation or comment.
Nice to get a quote from Debra Fischer about the new Alpha Cen discovery. I had thought her group actually had YEARS of data, at lower precision, with the 1.5 m CTIO telescope. Wish she had discussed that data also. If they really do have the precision of HARPS with their new spectrograph, they should be able to confirm this with one season’s worth of data, easily. This is, if I remember correctly, true given that this telescope is dedicated to this program while Alpha Cen is observable.
If there is an HZ planet and if subsequent imaging shows it has life, then wouldn’t that preclude anything but a robotic probe, or at best an orbiting only crewed mission?
What might be the limits to imaging, especially with a gravity lens?
Eriks writes:
My understanding is that communication via entanglement is ruled out, but I haven’t looked into the latest work on this, and I’m hoping some of the readers can weigh in on the matter — quantum entanglement is beyond my skill set.
I am all for the organic growth model instead of the limited one time shots when it comeses to working our way to the nearest stars, for several reasons:
A robot solarsail mission that does a flyby like you said takes 50 years, but the science it potentially could do from close by , as we speak is increasingly getting rivaled / equaled by contemporary near infrared telescopes here on earth and in direct orbit, soon we have a pretty decent look in all the nooks and crannies of space from the early epoch to direct exoplanet imageing.
A manned/settlement mission takes even longer and like you said we’d be better of testing the waters and practising living of the land here in our solar backyard first, in fact , thats why I believe that Elon Musks of doing something with Mars is an admirable goal, but i’d rather had seen an all out effort on getting full scale colonising/mining the moon done.
Brain Greene did Quantum Entaglement and telportation on his PBS series. As I understand it the entanglement is FTL but our use of it is limited by light because to sned information there has to be a tranmitter and reciever unlike the Star Trek transporter.
I like Andrews nano idea. I think for an initail nanao probe we should focus on what is needed for a flyby similar to the data we get on our interplantary .
Ideally if a sail it could be all incorporated into the sail iitself.
This all got me thinking about other research and activities that are going on that could be used for the interstellar effort.
Nano of course
The national ignition facility. NYT says little progress on fusion reactions BUT maybe something useful for engines and better yet -the NYT said the Laser work has DOD excited.
Space X says it can get to an asteroid and even Mars with humans in a decade .
That would be the precursor for buildling a laser array.
Of course accelerators for finding micro wormholes
Eriks and Paul: It is my understanding that quantum entanglement is perfectly well understood, and that the experts agree it does not provide a means for FTL communication. I believe this covers the experimental work for which the Nobel was given this year. I have heard of no claims that these experiments did anything other than confirm our understanding of quantum mechanics.
Andrew: I think that your notion of using gravity or atmosphere for deceleration does not work at the extreme velocities we are talking about.
Self replicating machinery would absolutely be a plus, the smaller the better, but as of now this is an area we know very little about. In addition to allowing for a small probe to build an observation platform of unlimited size at the target, it would also greatly facilitate the building of the immense lasers that would be needed for a light sail.
It is not clear how fast you can go with a light sail before it is eroded away by the interstellar gas. I see this as a huge problem, which to a lesser extent also applies to very small and thus unshieldable probes. At a good fraction of light speed, the interstellar gas turns into dense and highly energetic radiation.
Quantum entanglement can be used for communication, but not FTL communication. Although the change in the state of the entangled is basically FTL, it doesn’t give any meaningful information without other data from the source, which needs to be sent traditionally and thus no FTL. Or so the physicists say.
FTL communication would be *huge*; remember the neutrinos? Heuristic: if you can’t remember it clearly, or find it in 30 seconds online, it’s not real. Entangled-particle communications need a classical side channel to actually extract information. It’s a good way of communicating *securely* — the interceptable classical channel is meaningless by itself — not quickly.
As for a probe, Project Longshot seems the more ‘realistic’ fast engine we could do, and possibly fairly realistic. Uses a fission reactor for power, which forces fusion that provides thrust. No need to extra power from a fusion plasma. We’ve flown small fission reactors in space, mostly on Soviet satellites. Getting the mass of the reactor and fusion pulse unit low enough to be propellable, I dunno. Also getting the He3 for the chosen fusion reaction. But the plan imagined 4.5% c cruising speed, and stopping not a flyby. I’d guess doable if at all for $400 billion to $4 trillion, making it a matter of how badly we want to do it. Obviously not on a standard NASA budget. But at the low end, it costs one Iraq war. World GDP is about $60 trillion, or $600 trillion over a decade, so the high end is less than 1% of that. Unlikely to happen but you can dream.
(400 tons, mostly fuel, 4e5 kg. $20,000?/kg to GEO orbit, $8e9. Launch tends to be 20% of a satellite cost, so $40e9. Throw in another factor of 10-100 for development and you get my range. He3 is expensive at the moment so we don’t necessarily save money on most of the mass being fuel rather than electronics.)
I’d say we already have energy budgets big enough for century long missions, if we put fission reactors back up there; the tricks are converting the power to thrust efficiently, and building a system that can last a century.
I’d guess doable if at all for $400 billion to $4 trillion, making it a matter of how badly we want to do it. Obviously not on a standard NASA budget. But at the low end, it costs one Iraq war. ”
With such costs you could get projects like Exo-Earth Imager(actually I think it would cost below $400 billion)
http://www.oamp.fr/lise/seminaires/ESTEC2003.pdf
Which basically is far more efficient and cost productive than a probe. With probe you get some snapshots of planet up close, with such telescopes you get continuous imagery of exo-planets, and they are repairable, while probe can be lost.
Fly-by probes are nearly useless. You get to observe the whole system for about an hour, but individual planets will still just be small dots. Even if you could split off sub-probes for fly-by at individual planets, you would only get a few seconds close enough to each planet to make observations that could be characterized as “close-by”.
With rocket propulsion, the trade-off is really easy: With a given amount of fuel, if a fly-by mission takes time T, the same rocket with the same amount of fuel can stop at the destination if the trip is lengthened to 2T. In other words, if you want “instant” gratification in, say, 20 years, you get one hour worth of observation. If you are willing to wait 40 years instead, you get decades worth of observation. I think this is a fairly easy choice.
With a lightsail, things are not so easy. A stop at destination is probably just not feasible, or at least may become feasible only much, much later.
Nuclear rocket propulsion has very strict limits, but it is not a terrible option: The maximum possible effective exhaust velocity for a fission rocket is ~0.03 c, for fusion ~0.05 c. This is because of the limited amount of fuel mass that is converted to energy: ~0.1% for fission and ~0.25% for fusion. This makes for reasonable burnout velocities (available delta-V without obscene mass-ratios) of ~12% c and ~20% c, respectively. A trip to Alpha Centauri with a perfect fusion rocket would then take 20 years for a fly-by, 40 years for a full stop. Of course, in practice rocket performance will be a lot less than perfect, so your mileage may vary….
About the self-replicating machines: Obviously, those will only work with a full stop mission, so it is unlikely that they combine well with lightsails.
Several astronomers are already calling the discovery of this planet tentative and up for debate. The point of contention seems to be that an artificial signal may have been inadvertantly introduced into the data during the Herculean analysis. Many other confounding sources of noises had to be removed: a weak signal in a complicated date set does not sound like a “discovery” to me as this is being called in the media and elsewhere.
I am sure many of you remember Gleise 581g? I can see this very easily playing out the same way: other teams reanalyze the same data and the statistical significance lessens, other teams looking for the planet fail to find it, etc. I guess I am just not sure why the astronomers in question were not more cautious in the way they went about announcing this. Ideally, no “discovery” should have been announced until the other teams confirmed the planetary signal, as obviously D. Fischer’s team and the HARPS were in contact prior to the announcement and could have kept things under wraps until both teams found the same signal. I am guessing this did not happen because of competition to be the first to find planets in this system, but isn’t one of the cornerstones of good science supposed to be reproduction of results before momentous discoveries are announced to the public? One of the strengths of the Higgs-like particle discovery announcement last July was that two seperate teams of physicists using different instruments and indepedent analyses both confirmed the particle’s existence at the same mass.
Hopefully I am being too negative about all of this.
Eniac – IF you have that level of tech you might be able to cannibalise the lightsail to build a magsail while en route. Then you use it as a drag brake at the other end.
Now you can stop.
@ad: You are assuming that a magsail braking to a full stop is feasible. I do not think that has been established. I consider it unlikely, because, as I recall, the solar wind has only 1/10,000 the force of sunlight, and even light pressure is good at most for a few hundred km/s, a far cry from what is needed. Also, a lightsail is made of one material, a magsail of another. It is unlikely that one can be cannibalized for the other. The requirements are too different.
“With probe you get some snapshots of planet up close, with such telescopes you get continuous imagery of exo-planets”
Longshot is a stop probe, not a fly-by, it can be an orbiter and give you continuous imagery up close. Well, subject to crappy bandwidth, I guess. Make it more massive (or amortize that x100 development cost) and you could have it be a bus that drops off orbiters at the various planets. Or rovers! It can also make direct contact.
That said, I’m just raising awareness; if another commenter were pushing it, I’d probably be noting that a hypertelescope could easily have a much higher science/cost ratio, and have a much shorter payoff time. I don’t know how actually feasible it is, though. Including politically: we live in a world where NASA canceled the Terrestrial Planet Finder projects, interferometer and coronagraph both.
“You get to observe the whole system for about an hour”
Longer than that; realistic fly-by isn’t *that* fast, after all. Daedalus was supposed to be 12% c? Doubling Longshot’s cruise speed gives 9% c. 800 minutes to cross the 10 AU diameter of Jupiter’s orbit.
But yeah, seems low payoff for the effort. (Weren’t Niven’s wacky colony planets based on fly-by data?)
“spaceman October 20, 2012 at 16:32, Several astronomers are already calling the discovery of this planet tentative and up for debate.”
Agreed. Observations of this planet, if that is what it is, are right up to the limit of what we can currently hope to achieve from earth. I think people need to take a deep breath. Whole mountain ranges of conjecture are being raised from some molehills of data. Obviously, I respect this work and the people who did it, that part is real science and there is no shame if it doesn’t pan out. But if I were a betting man, I would pass on this one.
Note: I appreciate being educated on the difficulties in making observations of the Centauri system. Unfortunately, everything I have read regarding such observations makes me hesitant about this discovery.
The thing that throws me off about quantum entanglement is: what is the mechanism that causes the entangled particles to instantaneously signal a change in state?
I’m not a physicist, but it seems to me that the existence of such things makes the universe even more mysterious and not “well understood.” What is is about the structure of space-time that allows such two separated, but entangled particles to maintain a unique identifier with each other (in a sense) and be capable of affecting the other on the moment of “wave function collapse?”
It is almost as if the entangled particles are really just one inseparable thing.
spaceman said on October 20, 2012 at 16:32:
“I guess I am just not sure why the astronomers in question were not more cautious in the way they went about announcing this. Ideally, no “discovery” should have been announced until the other teams confirmed the planetary signal, as obviously D. Fischer’s team and the HARPS were in contact prior to the announcement and could have kept things under wraps until both teams found the same signal. I am guessing this did not happen because of competition to be the first to find planets in this system, but isn’t one of the cornerstones of good science supposed to be reproduction of results before momentous discoveries are announced to the public?”
LJK replies:
For the same reason that in 2010 NASA went ahead and declared they found microbes that lived on arsenic:
http://www.starshipnivan.com/blog/?p=6539
I am reminded of the book by Nietzsche titled Human, All Too Human.
@JoeP: there is a simple intuitive explanation for this. An entangled pair can be thought as two separate spins (think of them as vectors pointing in a direction), since they both are born from a single system with no spin, both spins must point in the reverse direction of the other to conserve angular momentum, but the actual direction in which they point is spread in various directions in quantum superposition.
Different observers at the far ends capture and measure the spins, but what happens is that the observer couples with the superposition of directions, and becomes himself a quantum superposition of observers seeing the spin is in multiple directions.
When both far away observers compare their measurement notes (over a traditional not-faster-than-light-speed channel) its where the magic happens; since physics laws demand that both observers agree that angular momentum was conserved, and since both observers are in superposition of multiple directions, each eigenstate of each observer will not interact with an arbitrary eigenstate of the other observers, but just the ones that globally conserve momentum. That means that each quantum state of spin direction measured of Joe will interact with the exact quantum state of spin direction measured by Alice such that both make a net angular momentum of zero
So in essence, there is no FTL action per se, the entanglement is only observable after the fact
JoeP: I was training to be a physicist before I switched gears, so I’m not one either, but I think your comments and questions are quite astute. We don’t understand the wave function collapse phenomenon all that well. It seems to be completely random within the overall probability envelope described by a particular event’s wave function. That’s the rub: for example, if I measure one of the entangled particles as spin up, I instantly know yours will be spin down, no matter how far apart we are. However, I can not (today at least) force my particle to measure as spin up only when I want to transmit a 1 and only spin down when I want to transmit a 0. I will see spin ups and spin downs happen randomly. So the random stream of spin ups and spin downs that you see, while perfectly complementing my measurements, does not convey information from me to you. As of today, anyway. We need a normal light-speed or slower channel to later compare notes and determine that the measurements do, in fact, complement each other.
But perhaps newer, better physics understanding someday will allow us to manipulate wave function collapse or othwise exploit it in a way that does convey FTL communication. I can hope, at least!
I also tend to think of entangled pairs as a “single inseparable thing”, as long as they are entangled, of course.
The irony about quantum entanglement is that the way most people understand it, it is really the same as a pair of notes: One with a cross and one with a circle, placed into sealed envelopes and carried far away in opposite directions. When you open one, you know instantly what is in the other. Yet, no quantum mechanics is involved. No FTL, either.
The people who understand what is really interesting and specific to quantum mechanics about the Einstein-Podolsky-Rosen argument (none who have commented here appear to be in that category) tend to be those with a fairly deep understanding of quantum physics. Those also happen to be the ones that understand why it has nothing to do with FTL communications.
Nevertheless, this is a fascinating subject matter, and has recently shown signs of potential future practical applications in the form of quantum computing. No FTL, unfortunately, but, just maybe, some nifty ultra-fast algorithms.
JoeP writes:
“I’m not a physicist, but it seems to me that the existence of such things makes the universe even more mysterious and not “well understood.” What is is about the structure of space-time that allows such two separated, but entangled particles to maintain a unique identifier with each other (in a sense) and be capable of affecting the other on the moment of “wave function collapse?”
I’m not a physicist either– a landscape architect actually– but physics fascinates me because I write fiction and because, well, it’s the only human effort actually looking at the nature of reality.
Fortunately, there are many books out there that explain, in a fair amount of detail and without going beyond calculus, exactly what is happening. I recently listened (while running on the beach) to Stephen Weinstein’s ‘Dreams of a Final Theory’ which, admittedly, is getting a bit old, having been published in 1994. But Weinstein captures the magic of physics in a truly unique way. His chapter on beauty and physics is a classic essay on beauty in general and physics in particular.
There are others. Many others. Brian Greene (especially ‘The Elegant Universe’) comes to mind, and I read a recent book about Richard Feynman that really helped me with high-energy physics but I cannot remember the title! Sometimes, Paul has a suggestion in his ‘Currently Reading’ icon over there on the right of the page (but his capabilities and training far exceed mine).
And the reason I’m off on this tangent is because these guys have given me the best and fullest understanding of modern particle and cosmological physics that a layman can hope to achieve.
My shelves are full of popular science writing. And the one thing that I learned from all of this is that our knowledge of reality is so pitiable, so infantile, that the future looks very bright indeed.
Eniac said on October 22, 2012 at 22:08:
“The people who understand what is really interesting and specific to quantum mechanics about the Einstein-Podolsky-Rosen argument (none who have commented here appear to be in that category) tend to be those with a fairly deep understanding of quantum physics. Those also happen to be the ones that understand why it has nothing to do with FTL communications.”
I have been assured by a message wormholed in from my doppelganger in an alternate universe that there my near duplicate understands quantum physics almost perfectly.
From the law firm of Einstein, Podolsky, and Rosen.
Eniac: Good point about people thinking of entangled particles as pre-written notes. To stretch your analogy a bit, the really strange thing about entanglement is that you may defer selecting an alphabet from which to select the symbols and have the notes written while the particles are already in flight.
For those who want to dig in deeper, here is a Wikipedia article on the EPR paradox and how it applies to entangled particle measurements: http://en.wikipedia.org/wiki/EPR_paradox#Measurements_on_an_entangled_state.
Eniac and Mike: My layman understanding of entanglement is that “pre-written” notes are a sort of hidden variable, which is a no-no.
Lets use the card analogy again. Lets say you have two cards that are entangled. The two cards must have complementary states: one black card, one white card.
Entanglement is not quite like putting each in an envelope while blindfolded (to prevent observation) and separating the two. When you observe one (say the black card), you know the other card must be white. And no “communication” occurs between the cards.
Rather, the idea (as I have always understood it) is that the states of each card are truly indeterminate. In other words, each card is not secretly black or white before being observed. They are not black or white. There is no value, you only know that you will get one of each should you observe one of them. If you go ahead and observe one of the cards, that causes “wave function collapse” and then the other card, no matter how far apart, is also resolved to the complementary state. Thus, when one of the card’s state is resolved (a black card), this is instantaneously communicated to the other card, forcing it from indeterminate state to a white card.
Be careful when saying “instantaneously communicated.” This is one of those areas where the lack of a theory which encompasses both quantum mechanics and spacetime physics (general relativity) can cause difficulties.
In relativity there is no such thing as a universal clock, therefore there is no absolute simultaneity. Something, somehow is communicated, or connected in some way. But it is premature to claim what spacetime curve (if any) connects the two measurements.
JoeP: I do not think you have correctly described the crux of the EPR argument. It is not about a difference between “unknown” and “truly indeterminate”. As far as I understand, there is no difference between those two. To measure simply means to get to know what was previously unknown, even in quantum mechanics.
However, in quantum mechanics, the “unknown” state is not described by probabilities, but by complex amplitudes. This has the effect that correlations between observers measuring different things about the same system come out funny, in the same way that electrons going through a double slit form “funny” interference patterns. I am afraid that I do not remember enough of my quantum physics to be able to explain it coherently, so Wikipedia might be a better bet for you to clear this up. Or, even better, you could try this beautiful work of true genius: http://www.scottaaronson.com/democritus/lec9.html. It will give you the distinct impression, which I hold true even though it is rarely expressed this way, that quantum theory is nothing less than a sweeping generalization of probability theory.
Anyway, the strangeness about the correlation between the two observers can only be apparent if you know about the state of both. This is only possible after waiting for a real signal to get from one observer to the other (or from both to a third), and so no actual FTL transmission of information has taken place, or ever could.
Eniac, you are probably right in that I should be more careful in choosing words.
Nevertheless, the crux of the matter is that there is an apparently real “action at a distance” that occurs FTL. This is what bothered Einstein. You are right that no useful information can be passed along using this mechanism. If you look at the link that Mike posted further up, I think that article explains it concisely. Here is is again:
http://en.wikipedia.org/wiki/EPR_paradox#Measurements_on_an_entangled_state.
And here is a very short and amusing video by Michio Kaku also discussing this:
http://www.youtube.com/watch?v=fp2jGxQlzko
Thank you for this really fascinating discussion. My layman’s introduction on QM was Manjit Kumar’s “Quantum” which is book I can wholeheartedly recommend – not only because of the detailed explanations on development of QM, but also for described debate between Einstein, Bohr, Schrodinger, Pauli and others.
Mentioned book describes EPR argument only briefly. However, it is clear that according to general understanding of QM concepts one cannot speak about pre – existing reality before the observation (at least according to Copenhagen interpretation of QM). So of course you cannot describe the entanglement as two different cards put in envelope and being sent to the opposite directions.
I have also understood that while the Copenhagen interpretation have so far satisfied all experiments there are people around looking for other solutions. Let us see if they suceed.
Ron S expresses my concern when he said “In relativity there is no such thing as a universal clock, therefore there is no absolute simultaneity. Something, somehow is communicated, or connected in some way. But it is premature to claim what spacetime curve (if any) connects the two measurements.”
In particular I feel very uneasy about quantum entangled particles in the very early universe, where GR indicates that it would seem easy for them to separate beyond their event horizons. I am lost as to what this could mean for the ultimate spacetime geometry in which QM and relativity are combined.
No, I think the crux of the matter is that there is NOT real action at a distance.
Whatever makes the entangled quantum pair different from the pair of notes does not change this aspect. Information is not transmitted from one observer to the other, but rather from the origin of the pair to each of the observers. The observers are linked by a common influence, not by cause and effect. To call this an “action” we would need the latter.
Eniac: Sounds what you are describing a hidden variable theory.
Like I mentioned earlier, there are problems with that interpretation.
After all, if what you are saying is that clear, then why was Einstein so put off by entanglement? There is no action at a distance and the particles simply have predetermined states. Obviously there are problems.
Rob: “…I feel very uneasy about quantum entangled particles in the very early universe, where GR indicates that it would seem easy for them to separate beyond their event horizons. I am lost as to what this could mean for the ultimate spacetime geometry in which QM and relativity are combined.”
I think you mean inflation rather than GR, since that is what supposedly “flattened” our observable universe.
In any case there probably isn’t anything particularly novel going on in the case you describe since it bears some (a lot of?) similarity to Hawking radiation where two opposite virtual particles go their separate ways rather than than recombining.
I suggest that they’re similar (but not the same) since in both cases the separation is due to spacetime curvature. Further, no individual observer is able, in either case, to inspect the states of both particles, or even to have a second observer communicate to the first observer the other particle’s measured state.
I’ll leave it to better minds than mine to discover what this might all mean.
I don’t think what I said implies hidden variables. What bothered Einstein was that if one particle’s momentum was measured, and the other’s position, both momentum and position would be known for both particles. The question here is known by whom? Each observer only knows one. You need the action at a distance only if you want to preserve the concept of a single reality, where each measurement uncovers the “one truth”. It is this notion that needs to be abandoned, not that of causality. I believe that is what Einstein found difficult.
What happens in a more formal sense is that by the act of measuring orthogonal quantities, the observers themselves become entangled, their combined wave function ceases to be a pure state. Thus it becomes possible for a particle’s momentum and position to both be known precisely without violating the uncertainty principle, as long as each is known to a different observer.
I think this leads to the much misunderstood Everett-Wheeler “many-worlds” interpretation of quantum mechanics. This designation is unfortunate, as it has nothing to do with “parallel universes” and all the non-sense usually associated with that. What is really at the heart of the matter is that reality is relative, i.e. it depends on your point of view. What we perceive as reality is just an arbitrary pointer into the universal wave function. Any other pointer would be just as valid.
Just an idea with regard to QE:
ok, so no instant information can be transmitted via QE between the two components of the couple, for the reasons clearly explained above.
However, how about if *several* QE couples (pairs) are used at the same time and their states are compared with each other. I mean in this case not a comparison of the two opposite ends of each pair, but of the next one at the same end, and the next, etc.
Wouldn’t this be comparable to some kind of binary code, 0’s and 1’s, or +’s and -‘s?
Hence wouldn’t this imply instantaneous information transfer after all?
Where is my reasoning going wrong?
The observation is key to what manifests to be non-local correlation (or as you put it, the “one truth”). And yes, that is the problem, and what requires an apparent FTL influence, because entangled particles that must maintain complementary states can only resolve to that state.
It is the act of observation itself that causes the state to be known, and that observation is required for both entangled particles to “commit” (for lack of a better word) to a particular state. It is not predetermined before observation/measurement.
This is the source of the apparent FTL phenom that I think you are not explaining away adequately. Just about everything I have read on this subject makes this point. Why would Michio Kahu intentionally deceive viewers of his blog video (that I linked above)?
JoeP: I don’t think Kahu wants to deceive anybody. This is just a subject on which people do not agree, and so different opinions are heard. If I understood your first paragraph correctly, you agree with me that it is either “one truth” or causality, but not both. I pick causality.
Even without EPR, you would get to this conclusion eventually simply by rejecting these untenable notions: that an observation somehow transcends normal physical interactions, or that there is a distinct boundary of some sort between quantum and classical systems, or that the collapse of a wave function is a real event rather than just a change in perspective. Action at a distance is just another one of these.
Let’s take 2 entangled particles.
Their wave-functions are not collapsed AKA they contain all possible states of the particles.
A measurement is performed.
The wave-function of a particle has collapsed. It now has a definite state.
The wave function of the other particle also collapses co a correlated definite state. This happens INSTANTANEOUSLY. What this means with respect to relativity and its relativity of simultaneity is still unclear.
This means that, philosophically speaking, information WAS transferred FTL.
Why can’t this be a FTL communicator? Because one cannot control the information sent; one is, effectively, sending noise FTL.
There is the rub: The wave function “collapses” only for the observer that does the measuring. The other observer does not know this and has no way of finding out. For them, the wave function still contains all possible states.
There would only be a problem if this “collapse” were an actual event. It is not. It is a change in perspective between the pre- and post-observation states of the observer, and remains local. During the observation, the pure state of the observer splits into a superposition of two states: That of having observed “spin up”, and one having observed “spin down”. For each, the wave function of the observed particle seems to have “collapsed” in accordance with their observation, but the combined wave function (particle+observer) has simply evolved according to the time evolution operator. There is no “collapse”, much less one that is instantaneous across a long distance or can be detected in any way.
As others have observed before, this statement is nonsensical in the context of relativity. There just is no such thing.
“There is the rub: The wave function “collapses” only for the observer that does the measuring. The other observer does not know this and has no way of finding out. For them, the wave function still contains all possible states.”
This is the measurement problem – it is THE unsolved problem of quantum mechanics.
Why? Because you have 2 contradictory perspectives – of the measured, collapsed wave function and of the not collapsed wave function.
One explanation for it is the many worlds interpretation, which you detail and support*:
“There would only be a problem if this “collapse” were an actual event. It is not. It is a change in perspective between the pre- and post-observation states of the observer, and remains local. During the observation, the pure state of the observer splits into a superposition of two states: That of having observed “spin up”, and one having observed “spin down”. For each, the wave function of the observed particle seems to have “collapsed” in accordance with their observation, but the combined wave function (particle+observer) has simply evolved according to the time evolution operator. There is no “collapse”, much less one that is instantaneous across a long distance or can be detected in any way.”
The many worlds interpretation solves the measurement problem by assuming that each time a wave function collapses, the entire universe splits in as many facets as there are possible outcomes of the wave function collapse.
This does solve the measurement problem but at a GIGANTIC cost – the universe multiplies to a ridiculous extent with every wave function collapse.
Philosophically, this breaks conservation of energy, of momentum, thermodynamics, etc, etc. And if there is even the slightest non-linearity between the newly created worlds, this breaking of fundamental laws is practical, as well.
Of course, the many worlds interpretation still does not solve entanglement satisfactorily:
From the perspective(S) – many of them – of the separated worlds corresponding to each possible outcome of the wave function collapse, information (noise, really) STILL was transferred FTL between the entangled particles.
From the larger perspective of the not collapsed wave function – it encompasses all the worlds – information was also transmitted FTL; the solutions of the wave function (corresponding to the worlds it encompasses) also show the FTL correlation between entangled particles.
*The many worlds interpretation is merely one of many interpretations of quantum mechanics; it was not proven to be correct and the others wrong – far from this.
“As others have observed before, this statement is nonsensical in the context of relativity. There just is no such thing.”
As I already said:
This happens INSTANTANEOUSLY.
Read here: What this means with respect to relativity and its relativity of simultaneity is still unclear.
When measured from the same special relativistic POV, the correlation between entangled particles IS instantaneous, surpassing the speed of light.
When measured from 2 special relativistic POVs – will this allow transfer of information (noise) into the past?
I doubt the experiment was performed.
If I were to speculate, I would say that this transmission of noise into the past would NOT break the Novikov self-consistency principle AKA causality would not be violated.