Talk:John von Neumann

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Former good article nominee John von Neumann was a Mathematics good articles nominee, but did not meet the good article criteria at the time. There are suggestions below for improving the article. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake.
September 16, 2006 Good article nominee Not listed

Nuclear tests and cancer[edit]

I really think this is extremely speculative, to say that his cancer was "possibly caused by exposure to radiation during his witnessing of atomic bomb tests.

Macrae apparently says: "It is plausible that in 1955 the then-fifty-one-year-old Johnny's cancer sprang from his attendance at the 1946 Bikini nuclear tests." Macrae, p. 231."

But this also seems wildly speculative. There's no evidence that he was exposed to significant radiation or fallout at Operation Crossroads. There's no reason to suspect it at all, in fact. He was not a grunt assigned to cleaning ships or dealing with exposed materials. Like other top-level personnel he was almost certainly well out of harms way and surrounded by copious radiation monitoring devices. His relationship to the test is quite different than the soldiers who were actually doing the dirty and dangerous work, the populations where significant correlations between their occupational hazards and their later pathologies have been painstakingly established. In general, though, Crossroads was not an especially dangerous test, a far as they went (epidemiologically speaking).

I find it really irresponsible when people assume that because a few high-level bomb people died of cancer, it must have been because of their work on the bomb. It's completely unscientific, as unscientific as pointing out that a number of bomb people lived well into their 90s (Hans Bethe and Edward Teller, for example, much less someone like Herbert York, who attended a huge number of nuclear tests over the course of his life and lived to be 88) implies that nuclear tests are safe. It is not at all clear this is a Marie Curie scenario.

If there hasn't actually been a strong connection implied, or a strong reason to think the testing is related to the disease (as is the case with genuine downwinders), I think such baseless speculation is inappropriate for Wikipedia, even if other authors have felt fine making such baseless and unsupported speculations in their own works. The appeal of this sort of story is obvious (the bomb-maker killed by his own developments) but lacking good reason to suspect it is true, I think we should opt on the side of conservatism and not try to make some sort of maudlin moral story out of his painful and rare disease. -- (talk) 21:02, 7 June 2009 (UTC)

That's a tricky problem. I believe the speculation is in several sources, but I am not familiar with John von Neumann biographies, just using the web. This is probably one of those cases where Wikipedia editors need to do a bit of original research before deciding what/how much to say.
For easy reference: We are talking about this edit. --Hans Adler (talk) 00:46, 8 June 2009 (UTC)
It is just as speculative of the editor at IP address to posit the statement given in rebuttal to Macrae's text. The remark regarding Herbert York is clearly a non-sequitur. We should not be basing article content on an example of poor philosophical reasoning. This objection aside, I agree that this point about origination of disease should be better corroborated. William R. Buckley (talk) 07:40, 8 June 2009 (UTC)
Wow, you've totally missed the point of what I said and what I took the time to be very careful about. Will you read it over again? My whole point was that such examples were equally meaningless, that they are not sound, that the mere presence of someone at a nuclear test and their later long life or early death was not causal. Ergo my saying it is "just as unscientific", etc. It is not "just as speculative" to say that there is no evidence given. -- (talk) 13:34, 8 June 2009 (UTC)
Here's a briefer summary for those that can't be bothered with multiple paragraphs: It's true that a biography or two has baselessly speculated about his attendance at a nuclear test as being his cause of death. And a good number of other biographies have parroted that speculation. But it's still just speculation, supported by no evidence. It really shouldn't be part of the article, even if others have made that speculation. There are good reasons to be suspicious of such speculations. -- (talk) 13:37, 8 June 2009 (UTC)
My prime concern was your use of a philosophical fallacy in your argument. Do a better job of describing your objection (such as by not relying upon fallacious reasoning) and I will not complain. William R. Buckley (talk) 00:42, 9 June 2009 (UTC)
While clearly it should not be stated in the article that his cancer was caused by experimental nuclear radiation exposure, the speculation deserves its due weight. Herman Goldstine also believed that von Neumann's (and all his other colleagues') cancer deaths were from such exposure. No single case of cancer has ever had its cause directly traced to a certain exposure event; there are always probabilities involved. That should not stop us from noting the coincidence, as many sources do, even if we do not imply the causation. Robert K S (talk) 21:29, 15 October 2009 (UTC)
I am told by Richard Gordon of the radiology department at U Manitoba that a correlation has been made, that radiography leads to cancer; the number of breast cancers later detected correlates linearly with the number of breasts previously tested for cancer in mammography. I will get the reference in specific, and promptly post that here. William R. Buckley (talk) 18:03, 6 December 2009 (UTC)
My two cents: this is both widely sourced and entirely speculative. Therefore it may be valid to repeat the claim but only as a report of speculation. Some wording along the lines of "It has been suggested in various sources that Von Neumann's cancer may have been caused by exposure to radiation during his witnessing of atomic bomb tests, though no specific evidence has emerged to support the theory." — Preceding unsigned comment added by (talk) 16:17, 28 August 2012 (UTC)

small editing suggestion[edit]

What does this mean: The public interest in this work was such that The New York Times ran a front page story, something which only Einstein had previously elicited. Nonsense, as written, of course, The NYTs has run lots of front page stories from its inception. Something is meant like ...only second scientific story on front page ( which seems dubious in any case). Suggest delete, edit or clarify. (talk) 14:34, 23 August 2009 (UTC)

This was an edit by User:Lunicornis. I will ask him about it. (Though he has not been active recently.) --Aleph4 (talk) 14:56, 23 August 2009 (UTC)

Not really true[edit]

With very few exceptions, all present-day home computers, microcomputers, minicomputers and mainframe computers use this single-memory computer architecture.

Yes, originally microcomputers and minicomputers used this architecture. But not now. Most are really a modified form of 'Harvard' architecture (Microcontrollers excepted). Yes, program & data are stored together in 'RAM', but the relation between the RAM and CPU nowadays is more like the relation between Magnetic drum/disk and CPU was back in the 80's. The instructions get moved to program cache, while data gets moved to data cache, two separate memories, and from there the system looks like a Harvard architecture. (talk) 03:25, 28 November 2009 (UTC)

Good luck with this one. There are a number of problems with this article, but some highly technical editors have tried in the past to get some accuracy in detail about computer architecture on wikipedia, but other editors won't allow it, no matter what your sources. Still, keep trying. --IP69.226.103.13 (talk) 04:38, 28 November 2009 (UTC)

The basic concept that von Neumann did invent was the fetch-and-execute cycle of an instruction pointer stepping through the same memory that holds the data. The reason for the common memory was that the access to instructions had to be as fast as the access to data, else one or the other would slow the machine down, hence why not use the same memory for both to save money? The Harvard machines were not fully electronic so the dual memory was a simplified way of saying they had electromechanical control units that were completely unlike the arithmetic units. Jfgrcar (talk) 05:42, 3 November 2011 (UTC)

I think it is not "not really true." I'm content with the wording in the article. I think a bible analogy is required here. Some people take it literally and others understand it through its meaning, derived or tradition. Here you have to understand the meaning which comes from Computation, the scientific sub-discipline, that Von Neumann essentially invented. I think we are getting caught up here in how a computer is physically produced; how industry implements it. A Turning machine or universal computer IS what EVERY computer is, regardless of architecture. It is the theory behind what makes a computer a computer. If you focus on the immediate architecture that Turing produced it was simply the LITERAL physical manifestation of his theory; the tool (technology) produced under the guidance of his theory. Harvard architecture is another technological implementation of a Turing machine. However, and this is where there is some confusion, all computing devices operate as if there is one address space from which the computing occurs. All computers are single memory in theory. Things like Paging is just another way to break "single memory" up while still presenting single memory to a computing device. It might seem trivial but you cannot break up the memory space of a computing device from its perspective or else computing fails.BinaryLust (talk) 05:57, 5 March 2012 (UTC)

Invention of the Computer[edit]

It is not correct to say that John von Neumann invented the computer. It is correct to say that he invented the concept of a stored program. Prior to this invention, computers were strictly data-flow machines, and programming was effected by rewiring of the machine. Please, know your topic before you edit corresponding Wikipedia articles. For complete information about the historical context of John von Neumann's work, see the book by Herman Heine Goldstine, titled "The Computer from Pascal to von Neumann." William R. Buckley (talk) 03:41, 15 October 2009 (UTC)

As Buckley well knows, it is even controversial to state that von Neumann invented the stored-program concept, and much ink has been devoted to this controversy. Robert K S (talk) 21:24, 15 October 2009 (UTC)
Agreed! I should have included the equivocating words *... reasonably correct to say ...* I imagine that you are referring most specifically to Zuse. William R. Buckley (talk) 21:42, 15 October 2009 (UTC)
Additionally, one could reasonably look at the Jacquard Loom and see an example of stored programming, however primitive. William R. Buckley (talk) 06:55, 18 October 2009 (UTC)
I'm not sure I've seen that argued before, and I think I might need to be persuaded of that. The primary idea in the stored-program concept is that data and instruction are stored in the same memory store. In the Jacquard Loom is there even a dichotomy of data and instruction? Aren't they identical (i.e., the data is the weaving pattern instruction to the loom)? Robert K S (talk) 07:02, 18 October 2009 (UTC)
The issue is that with data flow machines, the program is external to the device. With the Jacquard Loom, it is the data (thread) which is external to the device, and the program is contained in the sequence of cards used to control the loom. So, in a sense the Loom is the reverse case of the ENIAC. The nature of the memory is not the issue; just that programming was stored on the cards, instead of having to be physically implemented by the loom operator just before each pass of the shuttle. Is the suggestion more clear? William R. Buckley (talk) 00:26, 19 October 2009 (UTC)
More to the Moore School controversy, actually, but any complete discussion will mention Zuse as well. Zuse's ideas, of course, did not come to be known until after the stored program concept was already firmly established, and historians began looking into earlier claims. It is clear that it was the distribution of the First Draft, in combination with the Moore School Lectures, that cemented this architecture as the standard for general purpose computing machines in the '40s/'50s. Robert K S (talk) 04:11, 16 October 2009 (UTC)
Yes, there is a lot of controversy over the First Draft and the Moore School Lectures. William R. Buckley (talk) 04:53, 16 October 2009 (UTC)
The revert war now occurring between established editors and the person at IP address originates with a book, "The Spike: How Our Lives Are Being Transformed by Rapidly Advancing Technologies" by Damien Broderick. This claim by Broderick is ill-founded, being that it is unequivocal. As is generally recognised, von Neumann did not *invent* the computer. This recognition is not helped by the quote from Broderick, "Although von Neumann invented the computer we know today, ..." William R. Buckley (talk) 06:46, 17 October 2009 (UTC)
While I am better persuaded that von Neumann invented the notion of stored programming, I recognise that he is not the only person to understand the value of such programming. Hence, I do not insist on this position being represented in this article, as others view the evidence differently. I suggest that the person at IP address is relying upon a single reference, in spite of the many more publications which provide demonstration of claims (instead of assertion of claims), and that this behavior is inconsistent with the notion and goal of producing a quality encyclopedia. Please, expand your knowledge of the relevant literature, rather than persisting with an edit war. William R. Buckley (talk) 06:46, 17 October 2009 (UTC)
An important issue is apparent here. The anon has found a source which is verifiable (in the Wikipedia sense) to support his position. However, several editors above, as well as myself, think that the source contains information that is inaccurate and misleading. When two verifiable sources contain contradictory claims how is the matter to be decided? I suppose, as is happening here, by the weight of argument on the talk page. Possibly, the alternate versions need to be referenced, but in this case I think that is not needed as the anon's source is so marginal compared to the mainstream sources. Xxanthippe (talk) 22:15, 17 October 2009 (UTC).
It is not just that the anon sources (now two) are marginal. They also fail to cite a source for the conclusions upon which the anon relies, and so are clearly opinion. I do not argue that opinion is unimportant, even to an encyclopedia, but that such must be explicitly disclosed. William R. Buckley (talk) 06:53, 18 October 2009 (UTC)
There is a book that has been published by an author who interviewed Eckert (of Eniac fame) and who provided supporting evidence that von Neumann did not in fact invent the stored program computer, but stole the idea from Mauchley and Eckert. The story went that after (during?) building Eniac, they realized that they should have used the stored program architecture. They told of their ideas to vonNeumann, who was managing their work on the Eniac project, and vonNeuman quickly published a paper on it to ensure he would get credited for the idea. In other words, vonNeumann stole M&E's idea. (talk) 03:14, 28 November 2009 (UTC)
One should be cautious in accepting the assertion of an interviewee regarding claims. If indeed Eckert is correctly quoted in the book you mention, then he is in a conflict of interest position. Why should we accept that his (Eckert's) view of history is correct? After all, such a novel concept (as was then the stored program) would have been ripe for publication, and researchers are notoriously competitive in publishing novel concepts; i.e., no researcher likes to be scooped. William R. Buckley (talk) 18:15, 6 December 2009 (UTC)

To the editor at IP address, please, stop this edit war. The citations you give are not substantiated by other references, and constitute opinion, and nothing more; not one of these gives a citation for the claim. Mind you, I would (personally) be happy if the entire world accepted that von Neumann invented the computer. It is however not true, and this by review of documents which von Neumann himself wrote. William R. Buckley (talk) 05:01, 19 October 2009 (UTC) Again to the editor at IP address, you have cited a work that gives opinion. Indeed, in this case, the text is a blatant example of opinion, for why else would it begin with a plea, that "... we can all agree ..." Well, I for one don't agree. So, let me be a bit provocative with a counter example. I believe, without having asked, that the other editors of this page would be more inclined to suggest that, in the abstract, it was Turing who invented the computer. And, just to test my claim, will the other editors of this article please opine on this point? William R. Buckley (talk) 16:44, 19 October 2009 (UTC)

It should be obvious by now that the IP ignores this talk page, and just continues to push his text over and over again. It's pointless to make a discussion when one side refuses to participate. I think the time is ripe for other measures, such as semi-protection. — Emil J. 16:57, 19 October 2009 (UTC)
Yes, agreed, and at least we tried. I will continue to hold out the olive branch, in hopes that the IP will change his/her approach. William R. Buckley (talk) 20:01, 19 October 2009 (UTC)

Seemingly Misplaced Sentence[edit]

The sentence "he won the award from the U.S. Army for his creative thinking", which appears at the end of the chapter Politics and social affairs, seems out of place. Whaa? (talk) 10:45, 8 December 2009 (UTC)

Nationality in first sentence[edit]

If as has been asserted by one editor, von Neumann's notability started as a child prodigy, then the lead should state "Austro-Hungarian-American" rather than "Hungarian-American". The latter sounds like ethnicity is being emphasized rather than national citizenship. Yworo (talk) 09:08, 30 January 2010 (UTC)

The fact that his country was Austro-Hungarian Empire, and its declination after the World War I is not a reason to skip his nationallity. He was Hungarian Jew indeed. In Austro-Hungarian Empire like in Ottoman Empire, and other countries, which consisted of other countries, the nationallity was recognized according to the place of living. I do not like to emphasize his Hungarian nationallity, but many examples assert that the sole nationallity is more appropriate (see Gustav Mahler, or Bertha von Suttner; both were born in Austro-Hungarian Empire, but aren't considered as Austro-Hungarians). --Kiril Simeonovski (talk) 14:13, 30 January 2010 (UTC)
Birthplace and ethnicity/religion belong in the infobox and early life section. The information in the infobox is easily seen when starting to browse the article. It does not need to be repeated in the lead sentence. Yworo (talk) 14:36, 30 January 2010 (UTC)
No, it's not my point. The problem is Austro-Hungarian-American or just Hungarian-American. The examples above, clearly ote that, beside of the name of the country, the ethnicity is other. Another good example is Gustav Mahler. The text in the first paragraph is enough to conclude that John von Neumann became notable, and notable as Hungarian. Also, his main contributions in game theory were at the end of the 1920s. The notability gained before immigrating to the US was even greater than his notability in the United States. He moved as a keen mathematcian. He did not keep up his education in the United States. This occassion is same as with Enrico Fermi.--Kiril Simeonovski (talk) 18:39, 30 January 2010 (UTC)
P.S. If you think that only the last citizenship, or the citizenship at the time of dead is needed in the lead section, you got in incorrect. This makes sence, only if he has renounced his prevous citizenship, and his life spent in the prevous countries (the second is impossible).--Kiril Simeonovski (talk) 18:44, 30 January 2010 (UTC)
At the time period when he was naturalized, renunciation of previous citizenship was indeed required. I don't see what you have against Austro-Hungarian-American. On my talk page you yourself stated that his first notability was as a child prodigy during the years 1911-1913. Hungary did not then exist as a separate country, nor did it at his birth. Unless you are saying we need to use Austro-Hungarian-Hungarian-American. Is that what you want? Yworo (talk) 02:24, 31 January 2010 (UTC)
No! Hungarian-American is my propose. Austro-Hungarian is a little bit controverse, because Austria, and Hungary were separate, with own parliaments. It was like the European union today. Controversal is the Austro-Hungarian citizenship, which did not exist. Albert Einstein got an Austrian citizenship in 1911, which asserts that Austrian, and Hungarian citizenships were also separate. I gave examples above, and it's good for you to see. Then, why other notable people from Austro-Hungarian Empire are considered, only as Austrians, Hungarians, etc.?--Kiril Simeonovski (talk) 12:03, 31 January 2010 (UTC)
I agree with Hungarian-American. The other reasonable options are too complicated and only make sense if interpreted in one specific way. Ambiguity is a good thing so long as all reasonable interpretations are somewhat correct and the general idea is correct. Hans Adler 12:16, 31 January 2010 (UTC)
And if the reader is sufficiently sharp to recognise the attendant ambiguity. William R. Buckley (talk) 04:57, 13 April 2010 (UTC)

"Austro-Hungarian-American" is not a widely used term to represent ethnicity (I've never seen it before and I'm not even sure what it would mean), whereas Hungarian American is. In terms of ethnicity von Neumann was unambiguously Hungarian and unambiguously not Austrian. Robert K S (talk) 21:06, 13 April 2010 (UTC)

He was a Jew! This is Fact! — Preceding unsigned comment added by (talk) 11:36, 10 January 2012 (UTC)

von Neumann invented both the computer and the internet[edit]

at M.I.T. after he was brought over from germany. WTF it is a continuous source of abject humor to myself that no "academically official" organ on this planet TO THIS VERY DAY can willingly acknowledge either of these two facts! —Preceding unsigned comment added by Reetside (talkcontribs) 07:04, 8 May 2010 (UTC)

Apparently John von Neumann was involved in the development of an early form of weathe forecasting computer, known as "Maniac". An article in the 1954 edition of "Whitaker's Almanack" reads as follows: "MANIAC" WEATHER FORECASTS A calculating machine, designed to deal with the problems of accurate and rapid weather forecasting, has been developed at the Institute for Advanced Study, at Princeton, in New Jersey. The machine, nick-named "Maniac", (from mathematical analyser, numerical integrator and computer), was developed under the direction of Dr. John von Neumann. It contains about 2000 radio tubes and completes computation quicker than any earlier calculating machine. Scientists believe that the more accurate and rapid weather forecasting made possible by "Maniac", will be worth many millions of dollars to industries which depend on accurate prediction of weather, such as agriculture, shipping and aviation. If all factors which "make the weather", such as temperatures, barometric pressures, and wind velocities and directions, could be constantly measured at thousands of locations and many altitudes, weather predictions for (say), the next month, would be extremely accurate. The new machine provides a means by which the computation needed for accurate weather forecasting, can be done with sufficient speed to have the more detailed forecasts ready in time for practical use. Weather forecasters have been using equations which comprise great air masses, as isolated units, but new equations for "Maniac", comprise the whole atmosphere up to an altitude of 12 miles and deal with this air mass as if it were built up of imaginary "boxes", each with a volume of 240 cubic miles. These equations express the weather in each of the air boxes and then unite all the parts in a comprehensive equation. Before "Maniac", was built, there was no way of dealing with such a mass of figures quickly enough for use. Even the fastest pre-"Maniac" machines would take about two weeks to work out the answers. "Maniac" will solve the complicated equations in two and a half hours. Rjm111 (talk) 15:01, 7 July 2010 (UTC)

Sorry, but the comments above are not quite right. I hope they are not in the article! The computer project at the IAS produced a machine that did not have a formal name. Most people just call it the IAS computer. Several clones were built. Some were given whimsical names. The one at Los Alamos was named Maniac. The first computer weather forecast was made on the ENIAC because the IAS computer was not ready. The calculation was led by von Neumann and by Jule Charney. It was reported in a famous paper by Charney, Fjörtoft, and von Neumann in Tellus in 1950. Jfgrcar (talk) 02:33, 4 November 2011 (UTC)

Other Notable Hungarian Mathematicians[edit]

There are many other Notable Hungarian mathematicians. Paul Erdős was born in Budapest, Hungary on March 26, 1913 and was definitely in a league with Von N.

Suggestion to add his name to sentence

Even in Budapest, in the time that produced geniuses like Theodore von Kármán (b. 1881), Leó Szilárd (b. 1898), Eugene Wigner (b. 1902), and Edward Teller (b. 1908), his brilliance stood out.[5]

Respectfully, Steve M —Preceding unsigned comment added by (talk) 22:45, 26 January 2011 (UTC)

Functional analysis[edit]

I think that

"Von Neumann's functional-analytic program has dominated economic theory ever since"

should be explained. I cannot see any "domination" of economics by this. --Alex1011 (talk) 13:19, 13 May 2011 (UTC)

I rephrased the statement. (Look at the first chapter of Handbook of Economics (ME) for a discussion of convex sets. See also the essays on convexity in both editions of the New Palgrave, which are linked in the article Convexity in economics.)  Kiefer.Wolfowitz 22:19, 13 May 2011 (UTC)

Von Neumann models[edit]

We have Von Neumann model leading to computer architecture. But the economic growth models described in the text, sometimes called Von Neumann models, have no lemma. So again, the relevance of Von Neumann today in economics, for whatever, right or wrong, reasons, might be less than in the text indicated. --Alex1011 (talk) 13:31, 13 May 2011 (UTC)

I assume that you mean not "lemma" but "article on Wikipedia". See Dore et alia's book, which has Arrow, Samuelson, Harsanyi, etc. writing about von Neumann.  Kiefer.Wolfowitz 22:21, 13 May 2011 (UTC)
But that book was already cited. (I don't understand your comment: Wikipedia only covers a few topics, and is very bad in mathematical economics: I seem to be the only active person writing on these topics, at least lately.) Did I misrepresent the source cited?  Kiefer.Wolfowitz 22:24, 13 May 2011 (UTC)


In the formula, as it is now, λ is not the rate of growth, but the growth factor, 1 plus the rate of growth or 1 plus the rate of interest. Otherwise, I do not see 0 as result of the equation. I assume that A is the input-output-matrix and Bq the matrix of products produced (simple case Eq). --Alex1011 (talk) 07:54, 14 May 2011 (UTC)

Please correct lambda as the growth factor, and rename the interest accordingly.
For the complementarity condition, consider linear programming with positive data (as Schrijver noted) or (generalized) eigenvalue problems, or consult the literature cited.
The interpretation of A and B are different than Leontieff's model, so your assumption(s) need correction.
Best regards,  Kiefer.Wolfowitz 19:46, 14 May 2011 (UTC)

Sourcing to[edit]

Several details in the article are sourced to, which appears to be a self published website (apparently by User:E090: [1]), and almost certainly not a reliable source, especially for a WP:BLP. I've removed one of the details (here: [2]), but I'd like other editor input before I remove the source entirely. aprock (talk) 16:53, 27 September 2011 (UTC)

Von Neumann some of the first computer algorithms[edit]

"Von Neumann some of the first computer algorithms." this opening sentence under (science) is not grammatically correct, and the meaning is ambiguous. I am guessing that the words 'invented' or published were intended, in which case the statement is also making a very bold, *UNSOURCED* claim. I know the next 2 sentences give some additional context (with references), but they really just establish 2 algorithms he was instrumental in developing, it is still not established that these were some of the earliest invented/published/recognized computer algorithms. I'm rewriting that paragraph to remove the poor leading sentence, but keep the claims to the 2 algorithms. — Preceding unsigned comment added by (talk) 20:38, 29 October 2011 (UTC)

Photographic memory?[edit]

"Von Neumann had a photographic memory, and could recite exactly word for word any books he had read."

This little factoid sounds highly unikely. Any real evidence for this, other than an article from 1957? The wikipedia article on eidetic memory claims it has never been conclusively found to even exist. {{}}

The assertion is overstated. The otherwise awful biography by Norman Macrae "John Von Neumann: The Scientific Genius Who Pioneered the Modern Computer " does qualify this ability by saying (pages 8 and 152) that Von Neumann's ability to recite texts verbatim that he had read many years previously were restricted to those texts on which he had given specific concentration. — Myasuda (talk) 13:08, 2 November 2011 (UTC)

Von Neumann did have a photographic memory or "total recall" of text according to Herman Goldstine who tested von Neumann by asking him to recite "A Tale of Two Cities". Goldstine stopped him after 15 minutes. The story is reported by Goldstine, "The Computer from Pascal to von Neumann," page 167. Jfgrcar (talk) 05:30, 3 November 2011 (UTC)

I'm very well aware of that story. In fact, I own a copy of Goldstine's book. And Macrae refers to it specifically when he states that this ability was restricted to those texts on which he had given specific concentration. I don't have an objection to the term "photographic memory" but saying any books (as is done currently in the article) is an overstatement. — Myasuda (talk) 12:54, 3 November 2011 (UTC)

Goldstine says apparently any texts - and it's a claim repeated in quite a few sources. I think we just go with WP:V on this one. For debate about the possibility of these abilities in general, we have quite a lot of discussion on the eidetic memory article - but to bring it in here would be WP:Synth. Maybe we should link to the eidetic memory article? Avaya1 (talk) 20:56, 3 November 2011 (UTC)

No, you're wrong. Any books is your interpretation and is not what Goldstine states. Goldstine says "As far as I could tell, von Neumann was able on once reading a book or article to quote it back verbatim; moreover, he could do it years later without hesitation." Having the ability to do something is different from actually doing it in every instance. So the statement as you wrote it in the article ("could recite exactly word for word any books he had read") is not supported by Goldstine and should be modified accordingly. — Myasuda (talk) 00:59, 4 November 2011 (UTC)

I've re-phrased the sentence - although the difference in meaning seems to be slight. The sentence in the article says "could" and Goldstine says "was able", which have the same meaning.Avaya1 (talk) 15:50, 4 November 2011 (UTC)

I believe this discussion is losing track of the principles of attribution. As originally written, the article (and many others) was clearly wrong because it made a very general and unreferenced statement about von Neumann's memory. The proper thing to do, if you want to comment on von Neumann's mental powers, is to paraphrase what Goldstine wrote and cite him. It is then possible for the reader to look up the full reference and to decide for themselves. As for what type of photographic memory Johnny had, it is now impossible to say because he is dead and he was not examined by a specialist. Jfgrcar (talk) 02:17, 4 November 2011 (UTC)

I agree that he didn't have Photographic Memory in the popular sense. If I have one critique of this article it is the waxing poetic of his intellectual abilities. It is almost apocryphal. It even has a rabbi, err, teacher crying at the brilliance of his I love this guy as I'm a computer dude but it is a little too much without proper measurements. If the people around him thought that the mental puzzle in the article is a measure of intelligence or an example of his genius, then I find their own abilities to measure the guy suspect. I solved it "instantaneously" too. It is the equivalent of a medium level SAT problem, a level for above average put not particularly bright high school students.BinaryLust (talk) 06:06, 5 March 2012 (UTC)

You obviously didn't understand the point of the "fly puzzle" story. Most people don't need to have this anecdote explained to them, but here's some help: the story is illustrative of Von Neumann's mental speed not because he solved the problem "instantaneously", but because he solved it in an instant by constructing and then summing an infinite series. — Myasuda (talk) 04:39, 6 March 2012 (UTC)
Sure I did. Or rather I understand that that is one interpretation (which relates to the waxing poetic point I made and loathe). You just want to argue and contribute nothing. Fine then. You CLEARLY missed the humor in his response which is another interpretation of his answer which requires higher intelligence by the reader (you) and less waxing poetic. Sorry, but the guy wasn't autistic. This isn't a cartoon figure or a character in a movie. You can solve the problem with infinite series (which you also probably don't understand how he would have done it), but that is clearly overkill. He certainly KNEW this and like all quick witted people with people skills he was being coy (at least in this interpretation). Intelligence is about finding the most elegant, the most simple answer to a problem. Studies on intelligence show that the intelligent use LESS not more of their brain. They process more efficiently. He may have solved it this way out of habit, he may have been serious, but it was actually the DUMBER way to solve it, at face value. Ergo, this anecdote, like all anecdotes, is open to interpretation. To solve it like that was almost certainly out of nerd ego or a weird mathematical habit of his regarding infinite series that developed from his youth (possible and if so it uses Chunking not raw intelligence). Either way, he would understand that about himself.
Anyway, we don't even have to go there. You missed MY pointed. The point was based on the people assessing him. Initially, those assessing him, or anyone else, were merely measuring intelligence based on the speed of solving that problem, regardless of the manner. How he solved it moved him up a notch (infinite series is high school calculus; nothing particularly advanced either). My critique regards the almost apocryphal manner everyone in that section talks about him and I used this as an example to point out that those measuring him perhaps are most likely already in awe of his, clearly, great genius and so their assessment is suspect. He isn't a god and there are probably 50-100 people right now in Academia that are smarter than him. The talent in Science, but especially Math, has risen considerable over the past 50 years, pulling from virtually the entire population of the planet rather than just Europe/America.
OK, enough is enough, if you need me to spank you further, post on my page. I use YouTube to troll and vent steam, not Wikipedia. I respect Wikipedia too much. I suggest you do the same. Don't be jealous that I also was able to instantaneously solve the problem too although not with infinite Not all of us are cut out for math or have people skills, and sometimes both (like Neumann). Cheers!BinaryLust (talk) 05:44, 6 March 2012 (UTC)
I stand by my comment that you misunderstood the point of the anecdote -- I doubt anyone else here would disagree with that. Any now you are making unsupportable and incorrect aspersions about myself. And you are also continuing to display your ignorance of the people being quoted in the section "Cognitive and mnemonic abilities". You're free to give your opinion, but if you can't handle the responses in a grown up fashion, maybe you should find another place to spend your time. — Myasuda (talk) 13:50, 6 March 2012 (UTC)
You need to look in the mirror my friend. Instead of engaging me in conversation, you wished to insult my intelligence: "Most people don't need to have this anecdote explained to them, but here's some help," well before asking me for clarification. Even assuming you are right, you still "explained" it to me through insult. Educated people whenever they find something that contradicts their own initial thoughts say to themselves first, "Oh, what are alternate perspectives here," before making any type of critique, let alone an insult. What is more troubling, however, is that when I did give you my explanation, you simply denied my position (in my opinion you still don't understand my point) and reverted back to YOUR insulting prose: "you can't handle the responses in a grown up fashion, maybe you should find another place to spend your time." Frankly, I expect an apology. I wouldn't normally demand one, but you actually appear to be using your real name so this exchange here is representative of your character, which you choose to broadcast to the entire world, good or bad. Once again if you need more feedback use my personal page. It isn't right for other users to be distracted by this. BinaryLust (talk) 01:36, 7 March 2012 (UTC)
Yes, I suppose you're right that I should have phrased things differently initially. I apologize for responding to your original post in a prickly and inconsiderate manner. — Myasuda (talk) 02:10, 7 March 2012 (UTC)

Matthew Effect[edit]

The von Neumann pages are much better than when I first read them some years ago. Thanks to whoever is looking after them! Would someone please correct the misstatements about von Neumann under Matthew Effect? I once tried to remove the nasty reference to von Neumann, and encountered someone who always put them back. Jfgrcar (talk) 05:33, 3 November 2011 (UTC)

150 papers[edit]

Von Neumann wrote 150 published papers in his life; 60 in pure mathematics, 20 in physics, and 60 in applied mathematics.

60 + 60 + 20 = 140 + 10 ... what were the other 10 papers in? — Preceding unsigned comment added by (talk) 14:56, 10 June 2012 (UTC)

Other subjects. Yworo (talk) 23:33, 10 June 2012 (UTC)

Ferer not Szegő[edit]

I believe that von Neumann's tutor was Ferer, who became known for a theorem on the summability of Fourier series, and not Szego. Kiefer.Wolfowitz 12:42, 18 September 2012 (UTC)

Did you mean Lipót Fejér? That article contains this:

He was the thesis advisor of mathematicians such as John von Neumann, Paul Erdős, George Pólya and Pál Turán.

In any case the Wikipedia article on Gábor Szegő includes this:

Tutoring von Neumann: At the age of 15, the young John von Neumann, recognised as a mathematical prodigy, was sent to study advanced calculus under Szegő. On their first meeting, Szegő was so astounded by von Neumann's mathematical talent and speed that he was brought to tears.[1] Szegő subsequently visited the von Neumann house twice a week to tutor the child prodigy. Some of von Neumann's instant solutions to the problems in calculus posed by Szegő, sketched out with his father's stationery, are now on display at the von Neumann archive at Budapest.[2]
  1. ^ Impagliazzo, John; Glimm, James; Singer, Isadore Manuel The Legacy of John von Neumann, American Mathematical Society, 1990, ISBN 0-8218-4219-6.
  2. ^ John Von Neumann: The Scientific Genius Who Pioneered the Modern Computer, by Norman Macrae, American Mathematical Soc., 2000, page 70

So, it seems that Szegő was the tutor of a young von Neumann and that later Fejér was his thesis advisor at the University of Budapest. --Jeff Ogden (W163) (talk) 15:28, 18 September 2012 (UTC)

Death bed[edit]

"On his death bed, he entertained his brother with word-for-word recitations of the first few lines of each page of Goethe's Faust. Von Neumann died a year and a half later"

It doesn't make much sense to me to describe a deathbed 18 months prior to someone's death, deathbed usually implies the last few hours, or at a push days. My gut feeling is something has been mangled here, or a sentence has gone missing and 18 months refers to something else. Could someone more familiar with the material look at this. Thanks. MrWeeble Talk Brit tv 20:03, 9 October 2012 (UTC)

John von Neumann[edit]

A simple question here. If von Neumann was an Austro Hungarian American Jew how did he get the "von" on his name? That was traditionally an honorific assigned to nobility and indicated that the bearer was a Baron. This seems unlikely for a Jewish person given the anti semitic nature of pre WWII Europe regardless of his professional attainments. Of all the Jews who fled Europe, I don't recall any others bearing that honorific and I have noted here that the root article does not refer to Neumann as von Neumann until after his immigration to this country. Did he simply assume the name? That was not unknown at the time.

I must admit to being on shaky ground here with an assumption that might be basically incorrect, namely that a Jew would not be made a baron in any country in eastern Europe in the first half of the 20th century. — Preceding unsigned comment added by Tinnmann2 (talkcontribs) 04:28, 23 March 2013 (UTC)

Read the biography by McLean for details regard the acquisition of honorific title. William R. Buckley (talk) 18:29, 15 September 2013 (UTC)

Gödel originally considered his 2nd incompleteness theorem as "only a curiosity"?[edit]

I replaced the unreferenced sentence: "It is precisely this consequence which has attracted the most attention, even if Gödel originally considered it only a curiosity, and had derived it independently anyway (it is for this reason that the result is called Gödel's second theorem, without mention of von Neumann.)"

I have found no evidence that Gödel originally considered his second incompleteness theorem "only a curiosity." In fact, the opposite seems to be true. As Dawson states in his Logical Dilemmas: The Life and Work of Kurt Gödel (1st edition, p. 61): "The second of those problems, that of giving a finitary consistency proof for the axioms of analysis, was seen by Hilbert as the first step in his 'bootstrapping' program for securing the foundations of mathematics. It is not known when Gödel first turned his attention to the problem, but by the fall of 1930 he had found a solution that was profoundly unexpected." Dawson also quotes from a letter draft of Gödel: "The occasion for comparing truth and demonstrability was an attempt to give a relative model-theoretic consistency proof of analysis in arithmetic."

So Gödel was working on Hilbert's consistency problem when he discovered his first incompleteness theorem. When he discovered his second incompleteness theorem, it seems most likely that he would immediately appreciate its effect on Hilbert's program and not consider it as "only a curiosity."--RJGray (talk) 21:46, 16 April 2013 (UTC)

Fly Puzzle[edit]

The infinite series is 12 + 12*(1/5) + 12*(1/5)**2 + 12*(1/5)**3 + ... The series converges on 15. But the sum of the first four terms is 14.976. Therefore, although a formidable mental calculation, von Neumann could have reasonably guessed that 15 miles was the correct answer by mentally summing the first four terms, perhaps only the first three (14.88). (talk) —Preceding undated comment added 18:48, 28 December 2013 (UTC)

In general, if B is the speed of the bikes, F is the speed of the fly (F≥B), S is the initial separation of the bikes, and D is the total distance traveled by the fly, then

           D = S[F/(F+B)]{1 + (F-B)/(F+B) + [(F-B)/(F+B)]**2 + [(F-B)/(F+B)]**3 + - - - }

In our example,

     D = 20[15/(15+10)]{1+(15-10)/(15+10)+[(15-10)/(15+10)]**2+[(15-10)/(15+10)]**3 + - - -}
                 = 12(1 + 1/5 + 1/5**2 + 1/5**3 + - - -). 
                 = 12[1/(1-1/5)] = 12*5/4 = 15.

If B = 0, then D = infinity. If F = B, then D = S/2.

The first equation for D above can be written

            D = S[F/(F+B)]{1/[1-(F-B)/(F+B)]} = S[F/(F+B)][(F+B)/2B] = F(S/2B)
            D = 15(20/20) = 15 which of course is the easy way to solve the problem.

I suspect von Neumann was joking when he said he summed the series. (talk) 15:04, 5 November 2014 (UTC)

Hidden Variable No-Go Proof[edit]

I've edited this section based on a quote from von Neumann in the 1966 Bell paper, and the abstract of Bub's paper and its characterization in the Bell article. Based on the quote, there's no question that von Neumann portrayed his proof as ruling out all hidden variable theories, even though it in fact only rules out theories in which the hidden variables do not interact with one another (and therefore, it would seem, local hidden variable theories). It's also an error to say that Bell "misconstrued" the proof, since he was aware that David Bohm's 1952 hidden variable theory was grossly nonlocal, and it was his curiosity as to whether all hidden variable theories must be nonlocal that led directly to him developing his famous inequality. That suggests he was likely aware that von Neumann's error only applied to certain hidden variable theories. I also clarified what it was that von Neumann thought he was proving, and added the important other possible interpretation of Bell's Theorem: that nonlocality is real.

Now the question is: should Grete Hermann's obscure 1935 refutation (not discovered until 1974) also be mentioned? How about Einstein's awareness of the same flaw (cited by David Wick in The Infamous Boundary)? And how about the best refutation of them all, Bohm's 1952 theory that did precisely what von Neumann had said was impossible? Since it was that theory that led directly to Bell's proof, I think I may just go back and add it. — Preceding unsigned comment added by Emvan (talkcontribs) 15:25, 3 January 2014 (UTC)

According to Bub, the quote from von Neumann in Bell's 1966 paper, taken in the full context of von Neumann's proof, does not imply that all hidden variable theories are ruled out. However, the abstract of Bell's 1966 paper states: "The demonstrations of von Neumann and others, that quantum mechanics does not permit a hidden variable interpretation, are reconsidered." Bub also argues, based on the content of von Neumann's proof, that "Bell's observation could not have escaped von Neumann" -- in other words, von Neumann did not make the error Bell attributed to him. Hence, Bub considers Bell to have misconstrued von Neumann's proof.J-Wiki (talk) 03:14, 25 January 2014 (UTC)


I truly have a dislike of trying label people or retrospectively try to guess, speculate on somebodies inner most convictions. It still has to be said that some facts can be pieced togheter. Koldewe (talk) 18:51, 19 February 2014 (UTC)

Habilitation Berlin, Dec 1927[edit]

Excerpts of course list 1929/30 to 1932/33, Berlin

I plan to add the following article to the literature list:

  • Ulf Hashagen: Die Habilitation von John von Neumann an der Friedrich-Wilhelms-Universität in Berlin: Urteile über einen ungarisch-jüdischen Mathematiker in Deutschland im Jahr 1927, Historia Mathematica, Band 37, 2010, S. 242–280

and incorporate some conclusions into the part on Berlin/Hamburg, 1926–1932.

The habilitation was completed in mid of December, 1927 in Berlin. Because the habilitation was necessary for being a ′Privatdozent′, the first teaching term must have been the summer term 1928. In the scans shown on the right we see that von Neumann was on leave (′beurlaubt′) in the terms a) summer 1930, b) winter 31/32, c) winter 32/33, and that he taught courses in the other terms (with the exception of summer 1929). I also checked the course list of Hamburg for the summer term 1929 and found ′Axiomatik der Mengenlehre: Dr. v. Neumann. MoDiDoFr 16–17′. Therefore we can conclude that von Neumann was ′Privatdozent′ in Berlin from 1928 through 1932 (even 1933, but on leave), with one term in Hamburg (summer 1929) and two terms in Princeton (summer 1930, winter 31/32) before leaving Berlin. -- KurtSchwitters (talk) 14:17, 30 April 2015 (UTC)

I think the additional source
Ulf Hashagen: Johann Ludwig Neumann von Margitta (1903–1957). Part 1: Lehrjahre eines jüdischen Mathematikers während der Zeit der Weimarer Republik. Informatik-Spektrum (Springer), Vol. 29, 2006, p. 133–141, Part 2: Ein Privatdozent auf dem Weg von Berlin nach Princeton. Informatik-Spektrum, Vol. 29, S. 227–236.
would be helpful for the time until 1933. These are two articles in German containing a variety of source material (as does the other article by Hashagen on von Neumann's habilitation). Using part of the article I deleted the claim that von Neumann was the youngest Privatdozent at the Berlin university ever. --KurtSchwitters (talk) 15:04, 4 May 2015 (UTC)