On the scientific accuracy of “Avengers: Endgame”


Today Ben Lindbergh, a writer for The Ringer, put out an article about the scientific plausibility (!) of the time-travel sequences in the new “Avengers” movie. The article relied on two interviewees:

(1) David Deutsch, who confirmed that he has no idea what the “Deutsch proposition” mentioned by Tony Stark refers to but declined to comment further, and

(2) some quantum computing dude from UT Austin who had no similar scruples about spouting off on the movie.

To be clear, the UT Austin dude hadn’t even seen the movie, or any of the previous “Avengers” movies for that matter! He just watched the clips dealing with time travel. Yet Lindbergh still saw fit to introduce him as “a real-life [Tony] Stark without the vast fortune and fancy suit.” Hey, I’ll take it.

Anyway, if you’ve seen the movie, and/or you know Deutsch’s causal consistency proposal for quantum closed timelike curves, and you can do better than I did at trying to reconcile the two, feel free to take a stab in the comments.

46 Responses to “On the scientific accuracy of “Avengers: Endgame””

  1. RandomOracle Says:

    Let’s not forget that Tony Stark also describes the EPR paradox as “instead of going through time, time goes through you” *sighs heavily*. Personally I enjoyed the movie but the “science” was cringe-inducing.

    Scott, is there any sci-fi movie or tv show for which you enjoyed or at least weren’t offended by how it referenced real science?

  2. jeray2000 Says:

    I enjoyed the movie a lot. But I gave up on expecting the “science” to make any sort of sense since Ant-Man 2. As far as I’m concerned Pym Particles are equivalent to the magic Dr. Strange uses. As long as the movies are internally consistent I’m fine with it.

  3. Scott Says:

    RandomOracle #1: Absolutely, yes.

    Back in the day I loved Futurama, and for the past decade I’ve loved The Big Bang Theory and now Young Sheldon (though that’s not exactly sci-fi). It seems, consistently, like shows that play science for laughs do a better job with the science itself than shows that try to take it seriously.

    In terms of space movies, Contact, and more recently Passengers and The Martian, did very good jobs with the science. Though in terms of conveying the scientific worldview and attitude, the best I’ve seen might be the 1965 film The Flight of the Phoenix.

    Oh—in Jurassic Park, the character of Dr. Ian Malcolm, the “chaotician” who uses fractals and strange attractors to foresee that DINOSAURS WILL ATTACK, is so over-the-top stupid that I found it almost enjoyable. And while there are strong reasons to dislike Michael Crichton—still, if there’s such a thing as a clear “advance in science fiction,” then the concept of bioengineered dinosaurs, using DNA extracted from insects preserved in amber, should surely count.

    In terms of time-travel movies, I think I have to give the award for least cringeworthy to Austin Powers 2 and 3. 😀

  4. Job Says:

    One of the things i enjoyed about 2001 Space Odyssey was the minimal dialog.

  5. James Gallagher Says:

    Scott, I don’t think you should be posting spoilers for this movie just before its second weekend showing.

    It’s interesting when you do reference popular culture though, since you hardly ever talk about it, what were your musical interests as a youngster for example? Did you like The Cure? (who have just been indicted into the Rock & Roll Hall of Fame)

    I haven’t seen the movie yet (so thanks Scott) but FWIW I think time-travel backwards is just an entropy problem which can be solved locally by sufficiently powerful computers in the future, but time-travel into the future will be easy one-way for individuals who choose that

  6. Scott Says:

    James #5: Sorry about that!! I guess I just assumed that the time travel aspect must’ve been an integral part of the advertising for the movie, rather than a “spoiler.” As one example, this post was linking to an article published on a news site. Also, the writer explicitly told me that, if I cared, he wouldn’t be spoiling the movie for me by showing me the clips that I saw.

    Even so, immediately after you pointed this issue out, I added a spoiler alert to the top of the post.

  7. Scott Says:

    Also, James #5: I enjoyed Beatles and other classic rock as a kid—but that’s probably just because my parents were former hippies who constantly had it on. There’s a good deal of musical interest and talent in my family, but it completely skipped me.

  8. Harry Johnston Says:

    There’s been considerable discussion about the time-travel aspects of Endgame on scifi.stackexchange.com. Mostly pretty confused though.

    Someone did find a quote to confirm that Steve’s life with Peggy was in an alternate reality, not ours, which was perhaps the biggest point of contention.

  9. mjgeddes Says:

    I haven’t seen the latest, but I’ve seen previous Avengers and spin-offs. Very entertaining Scott, good stuff! They’re basically super-hero movies, so more like urban fantasy than sci-fi (Defiance of known laws of physics is built-into the genre).

    This 4-minute YouTube clip from a previous spin-off movie (Captain American: Winter Soldier) pretty much captures the essence of the characters and the whole Avenger universe. Watch this Scott. Great introduction to the character of ‘Captain America’ (Steve Rogers).

    In the clip, Captain America administers a beat-down of dozens of strong men in an elevator, showing super strength and immunity to electro-shock weapons. He then jumps out of a sky-scraper window, crashing through a glass ceiling and landing without scratch on the ground hundreds of meters below. Immediately springing to his feet and jumping on a motorbike, he then proceeds to fight and destroy a military aircraft! 😀

  10. Jalex Stark Says:

    Since this thread mentions depictions of science in fiction and mentions Futurama, I feel compelled to mention Keeler’s theorem on products of distinct transpositions, which is key to the plot of the Futurama episode “The Prisoner of Benda”.

  11. Shozab Says:

    Apparently Sean Carroll was hired as a scientific advisor for avengers. Supposing you were hired as a scientific consultant for a movie where they wanted to use complexity theory as a main plot point – which results would you ask them to focus on, in order to make the movie interesting / how would you like a movie focusing on complexity theory to play out?

  12. Scott Says:

    Shozab #11: There’s still the movie I dream of having made someday—the one whose climax involves the heroes running Grover’s algorithm on a million-qubit QC to break some cryptosystem, to learn some secret on which their victory depends, but the bad guys have surrounded their hideout, and now the good guys need to make a decision of whether to let Grover’s algorithm run to completion, or measure now and take their chances of learning the secret—with the understanding that if they don’t find it, then the state has collapsed and they have to start over from scratch (at least in this branch of the wavefunction!). This is an exciting plot that’s only possible because of the actual properties of a specific quantum algorithm, and that wouldn’t work with a classical algorithm. 🙂

    More broadly, though: there are many plots that it might take a person who knows a specific area of science to think of, and there are also many movies that require plausible scientific dialogue whose content is not important to the plot, and for those it makes a lot sense to hire scientists as consultants. What I have a fundamental problem with, is the common tactic of first inventing a sci-fi plot, and then asking scientists which sciencey jargon would render that plot plausible. That reminds me of the cargo cultists who Feynman imagined in his famous speech, who dealt with the event of no cargo arriving by wondering if they had designed their earphones with the wrong shapes. (“OK then, so what SHOULD Tony Stark mutter about the quantum realm as he switches on his time machine, in order for our movie to be way more scientific than Back to the Future?”)

  13. Joshua Zelinsky Says:

    Scott #3,

    Two recent scifi things I’d recommend are “Love, Death and Robots” which is a scifi anthology; they are a mixed bag. Some of them are very good though. There’s also a Youtube channel “Dust” which does scifi shorts some of which are very good.

  14. Rahul Says:

    Scott #3: “shows that play science for laughs do a better job with the science itself than shows that try to take it seriously.”

    I agree. My impression is that the funny-shows just do apt lateral references to deep science. People can grok the references on several levels based on the person. The serious shows get lost trying to explain it all.

  15. William Gasarch Says:

    My biggest objection to how science is used in movies is they go from basic research breakthrough to application in a few hours or even minutes. In a fictional universe Peter Shor would have, within hours of proving factoring is in Quantum-P, been kidnapped by criminals who force him to use it to have them break cryptosystems and steal money.

  16. Itai Bar-Natan Says:

    Regarding movies with good science, have you seen Contagion? It’s a realistic depiction of a catastrophic pandemic and how scientists and other institutions try to handle it. I enjoyed it.

  17. Scott Says:

    Itai #16: Haven’t seen it—thanks for the rec!

  18. JimV Says:

    I was very disappointed in the science abuse in “Endgame” and “Ant Man”. “Endgame” violates its own time-travel premise! (I guess I had better not say how; some of my nephews who can only go on Saturdays have found all the good seats reserved two weekends in a row now.) I mean, why can’t you have good characters and good dialog AND a plot that makes sense, all at once? I feel like I could have written a plot that left time-travel out and used Carol Danver’s linkage to the Space Gem to short-circuit Thanos’ Infinity Gauntlet and … . Granted, it wouldn’t have been scientific either, but it wouldn’t have pretended to be.

    Also, there was a little too much melodrama for my tastes. I can only give it a B-, whereas I gave “Captain Marvel” an A-. I have a feeling Marvel is going to go downhill now that Stan Lee is gone.

  19. Arch1 Says:

    Scott #3: I didn’t think of the credible-science angle when I saw “The Flight. of the Phoenix” as a kid, but one character did stick with me: A German engineer-type who, come what may, seemed to have what I later heard described (in reference to real life pro football coach Bud Grant) as a “stranglehold on reality.” Is he the character you mainly have in mind, or am I out in left field?

  20. Scott Says:

    Arch1 #19: Yes, I was an adolescent when I saw the movie, but I clearly remember it as the story of one socially inept guy, all substance and zero style, who somehow has to convince everyone else that he understands the reality of how to fix the plane.

  21. Anon Says:

    It’s The Ringer, not The Wringer.

  22. Scott Says:

    Anon #21: Thanks!! Fixed.

  23. Yovel Says:

    Scott #3- Have you seen twelve monkeys? It’s the best tine travelling movie I’ve seen (though I haven’t seen Austin Powers)

  24. Uncle Brad Says:

    The thing i *really* have a problem with is that ballista in GOT last night!

  25. Adam H. Says:

    My vote for time travel movie is “Primer.” Basically, a “Steve Jobs” and friend accidently invent a time machine in their garage while trying to launch a silicon valley tech company. No, it certaintly isn’t scientifically accurate in the sense that time travel into the past is likely impossible, but it concentrates on the difficulty of dealing with the paradoxes time travel would produce were it possible in a very realistic way. Ethical questions and the human toll take center stage over the technical stuff.

  26. Bill Kaminsky Says:

    I second Shozab’s comment at #11: Sean Carroll was indeed hired as a consultant for Avengers: Endgame. You can read a little about this here from this Jan 2019 article in Wired:


    If you don’t have time to read that short article, here’s the key passage:

    Being a science consultant in Hollywood also doesn’t come with a lot of power and prestige. “For one thing, you don’t get paid for it,” Carroll says. “You talk to the screenwriter or director or producer—whoever asked for your help—and you chat for a couple hours, and you do your best to give them advice, and then you never hear from them again.” … “I gave them great stuff,” [Carroll] says. “I hope they use it.”

  27. Scott Says:

    Shozab #11, Bill #26: Thanks! That must have been how the writers made the connection at all between time travel and David Deutsch (“Deutsch proposition”). I’ll ask Sean about it the next time I see him or talk to him. But I should make it clear that, given the way this process works (my own experiences give me zero reason to doubt Sean’s description), I don’t hold Sean responsible in any way for any forehead-bangers in the final product!

  28. Hamish Says:

    I was very pleased to hear Planck pronounced properly in the movie (“plank”, as a native German speaker would say).

  29. Bob Strauss Says:

    I think you can blame the original Star Trek for how science is used in TV and movies.

    In the heat of combat: “Scotty, can you polarize the neutrino beam to simulate a Klingon cloaking device?”

    (presses a couple buttons) “Already done, sir!”

    “Can you make the ship go 10 percent faster?”

    “Ach, maybe if you gave me three months and a couple of dozen engineers! I can’t perform miracles!”

  30. Dan Staley Says:

    I’m very disappointed that Endgame is not scientifically accurate about time travel, as opposed to the previous movies’ utterly-realistic treatments of Gamma rays, ARC reactors, Vibranium, and Pym particles.

  31. Gom Says:

    Bob Strauss at #29 – The original Star Trek didn’t really have all that much “technobabble”, it was Star Trek: The Next Generation and (especially) later iterations where it got ridiculous.

  32. Michael Says:

    Horrible news- there was a shooting at a STEM school in Colorado:

  33. Yaqub Says:

    The part of Avengers that had time-travel based on science was when Tony Stark met his father and almost forgot his bag before being reminded about it. If Tony Stark forgot his bag, then that would have created a bootstrap paradox where information from the future is given to his father, which ultimately benefits Tony Stark. Maybe the bootstrap paradox was avoided because of Novikov consistency principle.

  34. Jon K. Says:

    #12 That actually is a great plot for QM nerds! That movie should be made.

    One of the best science-based movie has to be “Digital Physics”. I recommend it for people who like Kurt, Alan, Claude, Andrey, and topics surrounding classical computing.

    In contrast with most Hollywood films, “Digital Physics” obscures its meaning and alienates its audience by focusing on topics most people aren’t familiar with or interested in. And there is nothing in the plot that makes you feel like something significant lies in the balance… besides the shelter and sanity of the main character who doesn’t do enough to win over the audience’s empathy… like I said, one of the best! 😉

  35. Jon K. Says:

    Oh, I also thought “Arrival” did a pretty good job of playing with the notion of time and mixing that into the core of the plot. (Or if that aspect was part of the short story the film was based on, I guess I should commend the author.)

  36. Itai Bar-Natan Says:

    @Jon K: My opinion of the movie “Arrival” is strongly biased by having first read “Story of Your Life” first. I can’t give you an impartial opinion on how it compares with other science fiction films, but definitely the way they play with the notion of time in the original short story is much more interesting than in the movie.

  37. تجهیزات کافی شاپ Says:

    The part of Avengers that had time-travel based on science was when Tony Stark met his father and almost forgot his bag before being reminded about it. If Tony Stark forgot his bag, then that would have created a bootstrap paradox where information from the future is given to his father, which ultimately benefits Tony Stark. Maybe the bootstrap paradox was avoided because of Novikov consistency principle.

  38. John Baez Says:

    I got interviewed by Business Insider about the physics of “Avengers: Endgame”. I tried to pour cold water on it but didn’t really succeed. I got off one nice quote about how neither the future nor the past is determined in quantum mechanics. The other quote of mine got bent out of shape by the context they provided.

    I regret that they didn’t use what I considered my best line:

    Q: At the film’s conclusion, Captain America goes back to put all the stones back in their right places, but then remains in the past to live out his life. There is significant discussion (including a disagreement between screenwriters and directors) whether that action triggered an alternate timeline, or whether he remained in the Marvel Universe’s timeline.

    Which of those theories is most accurate, in your opinion?

    A: It’s just a story! They can have it be whatever they want. This is like asking what Winnie the Pooh did after Christopher Robin grew up.

  39. asdf Says:

    Well, Andrew Granville (number theorist) and two co-authors are coming out with a graphic mystery novel about integers and permutations. It will probably be more mathematically accurate than The Avengers (dunno about Futurama):


    Mathematical discussion: https://dms.umontreal.ca/%7Eandrew/MSI/AnatomyForTheBook.pdf

    From blurb: “Prime Suspects is a graphic novel that takes you on a voyage of forensic discovery, exploring some of the most fundamental ideas in mathematics. Travel with Detective von Neumann as he leaves no clue unturned, from shepherds’ huts in the Pyrenees to secret societies in the cafés of Paris, from the hidden codes in the music of the stones to the grisly discoveries in Finite Fields. Tremble at the ferocity of the believers in deep and rigid abstraction. Feel the frustration—and the excitement—of our young heroine, Emmy Germain, as she blazes a trail for women in mathematical research and learns from Professor Gauss, the greatest forensic detective of them all.”

    Wow! 😉

  40. Jr Says:

    I agree it is cringeworthy when ludicrous sci-fi plots are given scientific justification, but the only people I blame are the scientists involved. Star Trek is a fun series, especially the 60ies version, but I have seen many people who think it took science seriously, which is a severe failure of science communication. I think it can be blamed on science popularisers desperate to make a connection between science and a pop cultural phenomena.

  41. Scott Says:

    John #38:

      This is like asking what Winnie the Pooh did after Christopher Robin grew up.

    Did you know that the new(ish) “Christopher Robin” movie explores precisely that question, or was your use of that example just a coincidence? 🙂

  42. Arko Bose Says:

    Even though Sean Carroll was hired as the (a?) scientific consultant for Avengers: Endgame, the final product clearly was not to his liking. The movie gives a nonsensical treatment to time travel, which I am certain caused him ulcers.

    The Avengers go back in time (leave aside the fact that once in a “quantum realm”, there is no way to deterministically end up at a specific point in spacetime) and retain the memories from their timeline’s future! If you are retaining a memory, the events in that memory belong in the past. So if you think you are changing the future consistent with those memories, that future belongs to a different timeline.

    For me, MCU’s best movie remains its first: Iron Man. The geekiest, nerdiest superhero movie ever made.

  43. Embarrasing Says:

    In https://arxiv.org/pdf/0910.4698.pdf you state R = {(0^n, y) : K (y) ≥ n} is in FBPP which is clear. Why is it not in FP^PH and where exactly is this problem?

    Also why is constructing an oracle A such that FBQP^A \not\subset FP^PH^A is trivial and not even related to quantum computing?

  44. Sniffnoy Says:

    Unrelated, but Scott, certificate’s expired again…

  45. RyanJL Says:

    If you like Jurassic park for “so bad it’s good” science, you may enjoy another comic book adaptation, CW’s The Flash. It’s like the science in Endgame but just non-stop. One of my favorites comes from an episode in the latest scene where The Flash and his crew fight a woman with super-powers which make everybody around her unlucky:

    “Hey, so we ran the sketches you made us through the CCP database, and Her name’s Becky Sharpe.
    She’s 24 years old, no priors.”

    “No priors? Then why is she in the database?”

    ” Because in the last three years she’s lived in Central City, she’s been rear-ended four times, her identity stolen twice A cat burglar literally burgled her cat.”

    “- [CHUCKLES] – Okay, you see, Harry? That is bad juju.”

    “Okay, Ramon, what you call juju, I call quantum entanglement.
    Discrete quantum particles are connected, and when triggered, they simulate a synchronicity that, to the untrained eye To the common eye Could seem like luck.”

    “So if Becky can affect the particles around her in a positive way, then their connected particles start spinning negatively.
    So when good things happen to her, bad things happen all around her.”

    “Great, she’s got a good luck field.”

    But Jeff Goldblum’s character will always be hard to beat.

  46. Shtetl-Optimized » Blog Archive » The quantum computer that knows all Says:

    […] weeks ago, a journalist named Ben Lindbergh—who’d previously asked me to comment on the scientific accuracy of Avengers: Endgame—asked me the same question about the miniseries Devs, which I hadn’t previously heard […]