Latest Paper on ArXiv
March 24th, 2007 by WaltI have seen references here and there to a new paper on ArXiv, The Riemann Hypothesis, by Tribikram Pati, that claims to derive a contradiction from the Riemann Hypothesis. I caution anyone who’s seen it to not to get too excited. The gap between thinking you have a proof and actually having a proof can be large, and it’s particularly large for a famous problem. Since ArXiv if for preprints and not-necessarily-polished work, people can and do upload papers that they have to retract later.
The argument in the paper is fairly elementary, using only basic real and complex analysis, so if anyone is interested in seeing if they can find a flaw in the argument I’d recommend giving it a try.
March 25th, 2007 at 3:36 am
Strange that Pati isn’t more careful in his statements in the paper. Eg why not write “I believe this is a correct proof” instead of confidently claiming it, especially after the recent Navier-Stokes fiasco? Perhaps the conventions of academic writing forbid it?
Surely his glory if he is right wouldn’t be affected, while in the more likely case that he is wrong the embarrasment might be smaller.
March 25th, 2007 at 6:06 am
Johan: Good question. When I skimmed through the paper, I was surprised at how poorly it was written, and how odd, say, the references are (using footnotes, which I don’t think I’ve seen in any paper in the last, say, 40 years). These to me were strong “crank” indicators, but apparently Pati is well-respected. More to the point, *if* I was ever in the position of thinking I’d proved a huge result like this (so, strictly in the hypothetical now) I’d spend soooo much time making the presentation perfect, and endlessly checking my results. Now, obviously anyone can make a mistake in the mathematics, but I don’t understand why one wouldn’t take more care with the presentation. At the very least, it makes the paper easier to peer review, and harder to dismiss as the work of a crank (compare De Branges) But then, as I say, I’m not that good a mathematician…
March 25th, 2007 at 9:30 am
“Apparently” Pati is well-respected? I’m having a hard time believing this. It’s anything at all but apparent to me. Do you have any links to people you’re sure aren’t sock puppets to vouch for him?
March 25th, 2007 at 9:31 am
Ooh, I can comment from this IP!
March 25th, 2007 at 12:21 pm
I’m not a sock-puppet. If you have access to MathSciNet, then you can do a search for Pati which brings up 44 papers. He doesn’t seem to have done much since the late 60s mind, but there was a conference held in his honor in Ujjain in 1999 (which is in India, and the proceedings were published by an India company).
March 26th, 2007 at 5:19 am
De Branges may be self-important and dramatic, but can you really call the man who proved the Bieberbach Conjecture a crank?
March 26th, 2007 at 11:33 am
Peter, no, one can’t. I should have phrased that better. I was trying to allude to how the mathematical world seems to have essentially just ignored De Branges’s claim that he’s proven the Riemann Hypothesis. It seems to me that this has happened because De Branges’s paper is essentially unreadable, and so, given his past history, everyone has just decided it’s not worth the massive effort to try to understand his proof. I say “decided”, but for example, I’m not personally even remotely in a position to read De Branges’s claimed proof. What confuses me is why De Branges hasn’t put more effort into making his paper easier to read, in which case, I think it would be much harder for experts to simple dismiss it. (Of course, arguably, then any errors would be easier to find as well, which gives one reason for writing unreadable papers).
March 26th, 2007 at 8:04 pm
Johan Richter:
Saying something like “I believe this is a correct proof” is unnecessary in a math paper. Of course the author(s) believe the proof is correct, or it would not be out there. You have to understand, mathematicians are impersonal and straightforward in their writing. Saying that one has disproved or proved something is being straightforward, not ‘overly confident.’ If someone finds an error, then the author(s) would have to correct it. It’s that simple. Just because this is a famous problem doesn’t mean the writing has to be more ‘humble.’ It should be written just like any other math paper.
The paper is quite readable to me, although I haven’t gone through it all. Putting references in footnotes is unusual, but it’s a minor stylistic issue and does not affect its readability.
April 2nd, 2007 at 2:37 am
It’s understandable from a psychological standpoint why the mathematics community is generally reluctant to try and check De Branges’ proof, based on his track record with submitting other proofs that turned out to not be true. But it seems strange to me that, going on three years now, there has not been SOME movement in that direction. Generally speaking, historically the first proof for a complex problem is normally difficult to follow, and then once it is verified, more readable ones tend to appear. I am actually somewhat mystified that somone or some group of individuals have not taken the time to make at least a preliminary search for fundamental mistakes in De Branges’ proof.
April 8th, 2007 at 11:57 am
The text at least looks like it was written by a professional, which
is more than can be said for most crank papers.
April 11th, 2007 at 6:27 am
Lemma 1 is correct.
April 16th, 2007 at 10:56 pm
All Lemmas are correct.
May 21st, 2007 at 1:31 pm
The proof is correct.
May 22nd, 2007 at 8:44 pm
[...] A couple of months ago I linked to Tribikram Pati’s purported proof that the Riemann Hypothesis was false. [...]
June 8th, 2007 at 1:01 am
I remove my last claim and allow myself to refer to some comments
of one of my colleagues at :
http://guests.mpim-bonn.mpg.de/kroetz/RH.pdf
June 9th, 2007 at 4:14 pm
[...] Bernhard Krötz has added another update on Tribikram Pati’s preprint. He now reports that the paper does not in fact disprove the Riemann hypothesis. Details provided by a colleague. [...]
June 10th, 2007 at 4:35 am
[...] The discussion of Pati’s paper was done by Bernhard Krötz, here. [...]