In which I play Phil Plait, because he's too busy to complain about popular entertainment these days

[mmcirvin]

Spoilers for Sherlock episode 3, "The Great Game":


This episode of Steven Moffat's modern-day Sherlock Holmes series (written by occasional Doctor Who writer and actor Mark Gatiss) carries over one of Arthur Conan Doyle's least plausible details about Holmes, that he's so concerned with keeping irrelevant facts from cluttering his head that he doesn't know the Earth goes around the Sun. Yet he remembers an astronomical detail when it is relevant to a case, which gives Watson an opportunity to say "told you so" but (as the Onion AV Club review pointed out) also raises the question of why he remembered that other fact in the first place.

But that detail didn't quite ring true for me for other reasons.

There's this "lost Vermeer" that Holmes knows has to be a fake, by his usual over-the-top "deductive" methods, but he can't prove it. Moriarty suddenly presents him with a ticking-bomb scenario in which he has to figure out the proof in a matter of seconds. What he realizes is that there's an extra star in the sky: the "Van Buren Supernova", which was only visible in 1858. So the painting couldn't possibly be a Vermeer. (In a previous scene, a hulking assassin has murdered an astronomy professor at a local planetarium, presumably to help cover up the fake.)

The Van Buren Supernova is fictional; in the real world, there was a supernova in the Andromeda Galaxy in 1885, but it only got up to sixth magnitude. Supernova 1987a in the Large Magellanic Cloud was a naked-eye object, but the last really bright one, in our own galaxy, was in 1604. Naked-eye novae do happen every so often. The name "Van Buren" was probably taken from the present-day astronomer Dave van Buren.

But suppose the supernova were real. Would this bit of the story make any sense? I don't think so. The problem is, it's too obvious.

There's a brief shot of the painting, and the supernova looks to be a very bright star, first- or second-magnitude, in the Winter Hexagon, near Orion. Maybe in Monoceros or Gemini, but far enough from the ecliptic that it would be unlikely to be a planet, though I'd have to watch closely again to be sure. Anyway, anyone with a passing acquaintance with the night sky would know that something was off, and if there really had been a supernova like that in 1858, they might well have known that too.

So it surely wouldn't have taken an astronomy professor, or Sherlock Holmes, to realize that the painting was a fake. Lots of art historians would have noticed it. That assassin in the planetarium would have had to bump off a lot of people.

And then there's the question of why someone setting out to fake a Vermeer would bother to paint a recognizable set of constellations in the sky, but also put that honking obvious supernova in there. This may have been inattention on my part, but it wasn't clear to me whether the fake was supposed to have been from 1858, or a modern fake (which would have made it even more baffling).

I was thinking the supernova was maybe intentionally put in there by Moriarty as part of his evil supergenius games, kind of like the way the Riddler's entire purpose in life seems to be to leave clues to get caught by Batman, but other dialogue seems to imply that the scheme was supposed to make sense until Moriarty decided to use it to toy with Holmes.

Comments

[mmcirvin.livejournal.com]

Those dates, just to be clear, are all for the arrival of light from the supernovae at Earth; the explosions in question actually happened far back along our past light-cone.

[ronebofh.livejournal.com]

You forge a painting from the original, not from a description of the original. Sheesh!

[mmcirvin.livejournal.com]

Well, unless the original doesn't actually exist any more, which is what I assumed was going on. But it still doesn't make a lot of sense.

[agent-mimi.livejournal.com]

Anyway, anyone with a passing acquaintance with the night sky would know that something was off,

Well, would people assume Vermeer was absolutely painting the truth, or just his version of the truth? Van Gogh's "Starry Night" is the example I'm thinking of here. I know VG and Vermeer didn't paint in the same style, but Vermeer did use exaggerated perspectives if I recall.

And then there's the question of why someone setting out to fake a Vermeer would bother to paint a recognizable set of constellations in the sky, but also put that honking obvious supernova in there.

That makes no sense at all.

[mmcirvin.livejournal.com]

That's possible, but then there would be the question of its similarity to the later event.

I'm assuming that the supernova would have been well-known, had it existed. The episode could have addressed some of these plausibility concerns by presenting it as a mere nova in a less conspicuous part of the sky (they could have even used a real one)--but that could have made it too implausible that even Sherlock could figure out the puzzle.

Did he say -IN- 1858?

[notr.livejournal.com]

I thought he was talking about a new object that had been visible since 1858, and it was just the description of it as a supernova that was wrong.

Re: Did he say -IN- 1858?

[mmcirvin.livejournal.com]

Wow, I never even thought of the idea that in this episode's universe, there's been an extra star there ever since 1858. That would make the plot more sensible at the cost of REALLY bad astronomy.

Re: Did he say -IN- 1858?

[notr.livejournal.com]

Are there no newly and persistently visible stars at all? Seems like the kind of thing that must have happened a few times--and that would be rare enough to stick in Holmes's strictly purged memory.

Re: Did he say -IN- 1858?

[mmcirvin.livejournal.com]

Star formation takes a pretty long time on human-historical timescales. It also tends to happen inside of nebulae, which can make it hard to see. I think that most main-sequence stars also take a while (thousands of years, at least) to blow up into giants when they start to run short on fuel.

There are variable stars that go through really weird, massive changes in brightness, and for all I know, some of them might have become persistently brighter after being inconspicuously dim for a long time. But I don't know of this happening to one of the small subset of stars that are bright enough to be naked-eye visible from Earth. There are several that become very dim and very bright periodically.

[agent-mimi.livejournal.com]

It makes you wonder if there was supposed to be some plot reveal or exposition explaining that Holmes had additional information that told him it was a fake painted in 1858.

Making the supernova large in the painting was probably just so those playing at home could see it on screen, although I agree that any time a TV show uses that kind of technique, it comes across as sloppy and unconvincing. There are so many other, better ways to convey clues.

[mmcirvin.livejournal.com]

He does have to poke around on his smartphone to get information about the "Van Buren Supernova", something I'd forgotten after seeing the episode (which also lends strength to Watson's chiding: he'd have gotten it faster if he hadn't had to look it up).

[chalumeau2.livejournal.com]

According to Mark Gatiss in the audio commentary for The Great Game, when they were first considering making the series into six 1 hour episodes, a forged painting a la the Vermeer would have been one such episode. I don't have details beyond that, beside the idea being recycled for TGG.

[pauldrye.livejournal.com]

My problem with the scenario was that I *always* try to figure out the constellations when a painting or photograph shows the night sky -- I'm just an astronomy nerd. I was about ten minutes ahead of Sherlock in this one, and that's just not right.

[sunburn.livejournal.com]

IIRC, Sixth magnitude is considered the limit of human-eye visibility, but someone in late 18th-century London would have a hard time spotting the full moon through all of that coal-fog. However, astronomy was in a sufficient state at the that period to notice and remark upon the sudden bright new star in a nearby "island universe."

[mmcirvin.livejournal.com]

Though until the 20th century, it was a minority opinion that "spiral nebulae" were external galaxies.

[sunburn.livejournal.com]

Hmm, evidently "Island Universe," goes back to Immanuel Kant in 1755, but the wikipedia article then uses the phrase in 2 more ways that suggest variously that it was an accepted concept of astronomers as being independent universes, followed by it being an accepted euphamism for a distance conglomeration of stars. Actually that last bit seems to be true, and also 20th Century.

At least Messier predates Holmes, so it's safe to say that "nebulous objects" could be found under the watchful eyes of astronomers.

The Lost Vemeer and van Buren Supernova

[(Anonymous)]

The only reason Holmes "remembered" anything about the van Buren Supernova was because it was mentioned in the planetarium where the Gollum killed the professor. As she is fastforwarding the video. He also asked John if he heard it too. That is the only reason he was able to deduce that it was the van Buren Supernova. He remembers things he hears when it's relevant, but discards the rest. It can be safely assumed that once that information was passed along, it was "deleted from his hard drive." As far as why someone would paint it in there was probably because a.)the painter wanted to be recognized (This is possible though unlikely.) and b.) If the lost Vermeer was truly lost, then he can only paint by description and was probably unintentional. or c.)Moriarty is the one who had the artist paint it in there so that Holmes could figure it out. That's the whole point of the episode anyway. Moriarty wanted to see Holmes shine, to see what he could do, so that he could set the plans into motion to beat him in Reichenbach. He basically wanted to know Sherlock's limits.