edds@unc.UUCP (Tom Edds) (12/05/86)
()
A fellow graduate student in our Lab (who is not on the net) has
exhausted local resources (i.e., several reference librarians) in
attempting to get a proper reference for the quote below. I thought perhaps
someone in net-land might be able to help, so I am posting this request
on his behalf. If you can help, please reply via e-mail to uncsmk@unc.BITNET
or via USnail to
Steven M. Kemp
L. L. Thurstone Laboratory
Psychology Dept.
Davie Hall (013A)
University of North Carolina
Chapel Hill, NC 27514
Any assistance will be considered a small miracle.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
re: Missing quotation, monkeys in the basement
"If a hundred chimpanzees were to be set before a hundred typewriters
typing for thousands of years at random. They would eventually
duplicate all of the works contained in the British Museum."
This is a paraphrase of one of several versions of a statement about
probability I have heard innumerable times. Newell and Simon referred
to it obliquely when naming the "British Museum" algorithm in 1954.
I have heard it credited to T. H. Huxley and to his son Julian Huxley,
but can find no reference to it or quotation of it anywhere.
Does anyone have a clue?
<posted by:>
Tom Edds UNC-CH Psychometrics
...mcnc!unc!edds
"Time flies like an arrow, but fruit flies like a banana."ram@nucsrl.UUCP (raman renukanthan) (12/30/86)
>>re: Missing quotation, monkeys in the basement >> >>"If a hundred chimpanzees were to be set before a hundred typewriters >>typing for thousands of years at random. They would eventually >>duplicate all of the works contained in the British Museum." >> >> Tom Edds UNC-CH Psychometrics >> ...mcnc!unc!edds >> >> "Time flies like an arrow, but fruit flies like a banana." I know for sure that Jeremy Campbell has quoted that in his book titled "The Grammatical Man". I longer have that book with me to look for references to that quote and also I am not sure if he has cited any references. Hope this helps. Renukanthan Raman Northwestern Univ. Comp. Sci. lab
tes@whuts.UUCP (01/01/87)
In article <3800001@nucsrl.UUCP>, ram@nucsrl.UUCP (raman renukanthan) writes: > >>"If a hundred chimpanzees were to be set before a hundred typewriters > >>typing for thousands of years at random. They would eventually > >>duplicate all of the works contained in the British Museum." A modified quote resulting in "all of William Shakepeares Works" was in either the book "How to Lie with Statistics" or its companion volume whose title I have forgotten. Both were 50's or very early 60's vintage books. Author long since forgotten. sorry for obscure reference, but I read these books 25 years ago. -- ----- Terry Sterkel -====---- AT&T Bell Laboratories --------- {harvard|allegra|ulysses|ihnp4}!whuts!tes ----- [opinions are obviously only my own]
jbn@glacier.STANFORD.EDU (John B. Nagle) (01/01/87)
Terry Sterkel claims that >A modified quote resulting in "all of William Shakepeares Works" >was in either the book "How to Lie with Statistics" or its >companion volume whose title I have forgotten. Both were 50's >or very early 60's vintage books. Author long since forgotten. >sorry for obscure reference, but I read these books 25 years ago. No, that quote is not from Darrell Huff's classic work, now in its 36th printing. Huff never discusses combinatorics. [REF] Huff, Derrell, "How To Lie With Statistics", New York, W. W. Norton and Company, 1954, ISBN 0-393-09426-X John Nagle
stefan@wheaton.UUCP (Stefan Brandle) (01/04/87)
In article <3800001@nucsrl.UUCP> ram@nucsrl.UUCP (raman renukanthan) writes: >>>re: Missing quotation, monkeys in the basement >>> >>>"If a hundred chimpanzees were to be set before a hundred typewriters >>>typing for thousands of years at random. They would eventually >>>duplicate all of the works contained in the British Museum." Does anyone seriously believe this? I don't make any pretense of being an expert on stats, but my little knowledge of combinatorics would indicate that the idea is crazy. Please set me right if I have missed something. -- -------------------------------------------------------------------------------- Stefan Brandle UUCP: ihnp4!wheaton!stefan Wheaton College "But I never claimed to be sane!" ---------------------------------------------- MA Bell: (312) 260-4992 ---------
tes@whuts.UUCP (01/05/87)
In article <13880@glacier.STANFORD.EDU>, jbn@glacier.UUCP writes: > Terry Sterkel claims that > >A modified quote resulting in "all of William Shakepeares Works" > >was in either the book "How to Lie with Statistics" or its > >companion volume whose title I have forgotten. > >sorry for obscure reference, but I read these books 25 years ago. > > No, that quote is not from Darrell Huff's classic work, > now in its 36th printing. Huff never discusses combinatorics. > What about the 1959 edition? Also, do you have a reference for the "companion volume". -- ----- Terry Sterkel -====---- AT&T Bell Laboratories --------- {harvard|allegra|ulysses|ihnp4}!whuts!tes ----- [opinions are obviously only my own]
mmar@sphinx.UUCP (01/06/87)
In article <372@wheaton.UUCP> stefan@wheaton.UUCP (Stefan Brandle) writes: >In article <3800001@nucsrl.UUCP> ram@nucsrl.UUCP (raman renukanthan) writes: >>>>re: Missing quotation, monkeys in the basement >>>> >>>>"If a hundred chimpanzees were to be set before a hundred typewriters >>>>typing for thousands of years at random. They would eventually >>>>duplicate all of the works contained in the British Museum." > >Does anyone seriously believe this? I don't make any pretense of being >an expert on stats, but my little knowledge of combinatorics would >indicate that the idea is crazy. > >Please set me right if I have missed something. > >Stefan Brandle UUCP: ihnp4!wheaton!stefan The quotation doesn't seem sure how much time it's allowing. If "thousands of years" means 2000 < T < 1,000,000 (otherwise it would be "a million" or "millions"), then it also strikes me as crazy (noting that it only calls for 100 chimps). But if we ignore the "thousands", or suppose that any large number counts as "thousands", then emphasizing _eventually_ gives us unbounded time, and it becomes not so crazy. In his book _One, Two, Three...Infinity_, George Gamow does an approximate calculation of the number of combinations available for a single medium-length printed line (I don't recall the details), and comes up with something huge which he analogizes to all the atoms in the accessible universe (red-shift boundary) computing it in parallel without duplication, at the rate of atomic vibrations, ever since the big bang, they would not yet have made a sizeable dent in the problem. The last time this came up on the net, people pointed out literary etc allusions to the problem. My favorite is a Bob Newhart routine (on one of his early phono albums), in which he imagines that someone would have to be hired to check the chimps' output. The ending is roughly this: Hey, Harry, I think this one here has come up with something. "To.. be.. or .. not .. to.. be.. that.. is.. the..... gzornenplatz.." -- -- Mitch Marks @ UChicago ...ihnp4!gargoyle!sphinx!mmar
lew@ihlpa.UUCP (Lew Mammel, Jr.) (01/06/87)
> The last time this came up on the net, people pointed out literary etc > allusions to the problem. My favorite is a Bob Newhart routine (on one of > his early phono albums), in which he imagines that someone would have to be > hired to check the chimps' output. The ending is roughly this: > Hey, Harry, I think this one here has come up with > something. "To.. be.. or .. not .. to.. be.. that.. > is.. the..... gzornenplatz.." > -- > -- Mitch Marks @ UChicago I like that one too, but my favorite is from Stanley Myron Handleman. ( He portrayed a nebbish Brooklynite character and seems to have had a very brief career as a comedian about 12 or 15 years ago. ) He told the following story: I heard that if you put a bunch of chimpanzees in a room with typewriters they would eventually type all the works of Shakespeare so I decided to try it. I got a bunch of chimpanzees and put them in there and after a couple of days I thought I'd check on them and see if they came up with anything. But when I looked in there [shocked tone of voice] you know what? They weren't typing! They were just messing around in there! Lew Mammel, Jr.
csma@lifia.UUCP (Christian de Sainte Marie) (01/14/87)
>In article <3800001@nucsrl.UUCP>, ram@nucsrl.UUCP (raman renukanthan) writes: >> >>"If a hundred chimpanzees were to be set before a hundred typewriters >> >>typing for thousands of years at random. They would eventually >> >>duplicate all of the works contained in the British Museum." I know of a slightly modified quote: "If you put a chimp before a typewriter and if he types randomly and the output is the first chapter of "Therese Desqueyroux", you made an error: it is not a chimpanzee, it is Francois Mauriac" (from "the little encyclopedy to comfort the good-for-nothings" by Cavana) PS: does this discussion really belongs to comp.ai? -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Ch. de Sainte Marie - LIFIA/IMAG - BP 68 - 38402 St Martin d'Heres - FRANCE csma@lifia.UUCP {seismo.css.gov!mcvax!inria!imag}!lifia!csma
marty1@houem.UUCP (01/16/87)
In article <1281@lifia.UUCP>, csma@lifia.UUCP (Christian de Sainte Marie) writes: > >In article <3800001@nucsrl.UUCP>, ram@nucsrl.UUCP (raman renukanthan) writes: > >> >>"If a hundred chimpanzees were to be set before a hundred typewriters > >> >>typing for thousands of years at random. They would eventually > >> >>duplicate all of the works contained in the British Museum." > .... > PS: does this discussion really belongs to comp.ai? It has already been established by previous discussion that in terms of probability theory the chimps will eventually succeed, but it will take longer than various limits, including the expected life of the universe. If the claim of the chimps were true, it would say that any stupid machine can duplicate all the creativity of humans, and all it takes is incremental improvement in the machines to do it faster and better. But if the chimps' claim is not true, and in practical terms it is not, then there is a qualitative difference between a bunch of chimps and a bunch of people, and we have to work really hard to come close to making a machine that does what people do. Put another way: the hypothesis that all the works in the British Museum could have been created by chance is rejected. But I admit that the source of the quotation is a side issue. M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houem!marty1
mendozag@pur-ee.UUCP (01/16/87)
In article <1281@lifia.UUCP> csma@lifia.UUCP (Christian de Sainte Marie) writes: >>In article <3800001@nucsrl.UUCP>, ram@nucsrl.UUCP (raman renukanthan) writes: >>> >>"If a hundred chimpanzees were to be set before a hundred typewriters >>> >>typing for thousands of years at random. They would eventually >>> >>duplicate all of the works contained in the British Museum." > >I know of a slightly modified quote: >"If you put a chimp before a typewriter and if he types randomly and the >output is the first chapter of "Therese Desqueyroux", you made an error: >it is not a chimpanzee, it is Francois Mauriac" >(from "the little encyclopedy to comfort the good-for-nothings" by > Cavana) > >Ch. de Sainte Marie - LIFIA/IMAG - BP 68 - 38402 St Martin d'Heres - FRANCE > >PS: does this discussion really belongs to comp.ai? Feel guilty now? huh? :-) Now for some serious discussion... A few years ago I read a book with a good discussion on the "Monkeys at the Typewriters" problem. I don't recall the exact reference and the library's computerized retrieval system is down for a few days so I cannot check. It is (was?) published by Prentice-Hall and written by some physicist from Yale with last name Bennett (I think). The book deals with some scientific programming (with BASIC, but don't disregard it because of that) and it has a chapter on Language (about chap. 4) where discusses the problem very nicely (it even mentions Bob Newhart's act). It seems that the problem has been popular for some time especially since some Sir Arthur Eddington (around 1927) used it on some statistical mechanics lectures. (I first heard about this problem with the name "Sir Arthur Eddington Monkey and Typewriters Problem"). The book also deals with relations of this problem to information theory. I might have copies of that section but no references since at the time I was not interested in references and such :-(. Victor M Grado pur-ee!mendozag mendozag@ecn.purdue.edu
rlw@briar.UUCP (Richard Wexelblat) (01/20/87)
``There is a special department of hell for students of probability. In this
department there are many typewriters and many monkeys. Every time that a
monkey walks on a typewriter, it types by chance one of Shakespeare's sonnets.''
Bertrand Russell, @I{Nightmares of Eminent Persons}ray@bcsaic.UUCP (01/23/87)
> In article <1281@lifia.UUCP> csma@lifia.UUCP (Christian de Sainte Marie) writes: > >>In article <3800001@nucsrl.UUCP>, ram@nucsrl.UUCP (raman renukanthan) writes: > >>> >>"If a hundred chimpanzees were to be set before a hundred typewriters > >>> >>typing for thousands of years at random. They would eventually > >>> >>duplicate all of the works contained in the British Museum." I just yesterday stumbled over a reference which may indicate you"re not looking far enough back. In a paper arguing for "creationism" the writer says "In the Huxley-Wilberforce debate over Darwinism, Huxley claimed that given "enough" time, an Earth full of monkeys could type the works of Shakespeare." This would be T. H. Huxley, in the 1860's, I believe, and the context was an argument whether life could have come to exist on this planet "by chance". People (as evidenced by the existence of this paper) are still arguing four generations later. Happy hunting! Ray