Talk:Marian Rejewski

Missing image Key-crypto-sideways.png WikiProject on Cryptography

This article is part of WikiProject Cryptography, an attempt to build a comprehensive and detailed guide to cryptography in the Wikipedia. If you would like to participate, you can choose to edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.

Contents

1 Cryptographer or Cryptologist? 2 Single most valuable contribution? 3 Some biases 4 Authoritative account 5 Extended Rejewski quote

Cryptographer or Cryptologist?

Why cryptographer instead of cryptologist ? Wojsyl 20:12, 29 Apr 2004 (UTC)

Wojsyl,

There has been some considerable debate on this here on WP, and the concensus was for cryptography, cryptographer and so on. See Wikiproject Cryptography, List of topics in Cryptography, the article crytography, etc. I made the change to fit with usage on WP. No other reason. ww 23:46, 29 Apr 2004 (UTC)

Frankly, he was a cryptoanalyst rather than cryptographer. The difference is similar to this between military intelligence and counter-intelligence. Wojsyl 12:16, 24 Jun 2004 (UTC)

I didn't know that we had decided on this for use in article text; I know we decided to merge cryptology and cryptography as articles, and use "cryptography" in article names, but I personally don't have a problem with "cryptologist" in the context of historical cryptography where the word gets used more frequently. — Matt 13:03, 24 Jun 2004 (UTC)

Matt, This was before you became active. The problem was that everything could fit under cryptology AND cryptography (and the variants of same) and in the interest of lessened confusion... It's a minor point really, somewhat like y v i I suppose. But no one spoke up for cryptologist, so... ww 20:47, 24 Jun 2004 (UTC)

I believe that cryptology consists of two branches: cryptography and cryptanalysis, at least in Polish it does. Is it the same in English ?

Not really. Cryptology and cryptography are congruent in meaning in English. ww

So there's nothing wrong in calling cryptographer a cryptologist, but cryptanalyst is not a cryptographer (while he is still a cryptologist). Does it make sense ?

A confused sense in English. Polish and English appear to be somewhat skew in the categorizations the language uses. ww

I would therefore change it back to cryptologist here unless there are strong objections. Wojsyl 04:01, 25 Jun 2004 (UTC)

I don't object. There's really two styles that you encounter in English. The first is to treat the entire field as "cryptology", split into two branches: "cryptography" and "cryptanalysis". However, more recently people have been calling the entire field "cryptography" instead. It seems that on Wikipedia we have chosen to use "cryptography" in article names, presumably because the second style is more popular. I think we should avoid "cryptographer" (as it may be misleading in the first style) and use one of "cryptologist", "cryptanalyst" or "codebreaker". — Matt 14:59, 25 Jun 2004 (UTC)

M&W, English has a very large vocabulary and there are several (many) ways to say most anything. In other languages (perhaps Polish?) there is less confusing profusion of choice, and so less dithering. At least one can hope others avoid this occasional paralysis. My understanding of the history in English is that cryptography (and presumably cryptographer -- as one who does cryptography) was the original term; probably chosen in some of the early works on the subject on the Continent; English writers, though not English cryptanalysts and perhaps cryptographers (eg, Thomas Phelippes (sp?), John Wallis, ...) who were fully au courant), were somewhat late into print in comparison. Cryptologist is, as I understand it a later adoption (backforming some uniformity with such things as archaeologist (BP spelling here?) and, with what I think to be one of singularly ugly words in English, theologist; theologian is so much better!). At base, it's a stylistic choice. The -ologist forms have a somewhat unpleasant (I can't find a word for this so I'll misuse one) flavour when being said, at least to this English speaker (albeit of a corrupted colonial form of the language), and a non English sound as well to this English speaker's ear. Something Germanic in the sound sequence somewhere or something.

I personally prefer the following structure for English words in this field and use it in my writing on the subject. The entire field is cryptography, which includes cypher and code design and implementation (done by cryptographers as I don't know a better term for those who perform this function (ie, these functions as they are not congruent in re personnel), and -ologist as an alternative to cryptographer is ugly and unacceptable aesthetically (BP spelling again?)), and cryptanalysis which is done by cryptanalysts. Popular usage mangles the meaning of the latter and takes all cryptanalysis to be codebreaking (including Kahn's book title, unfortunately -- it was probably a tyrannical editor or some publisher's marketing droid who insisted) and those who do it as codebreakers.

I'm curious, regardless of how one expects it to be done in English, about the usages in Polish. Can you enlighten we who are innocent of Polish (except in this case that I know you have more letters than we do)? Thanks. ww 15:49, 26 Jun 2004 (UTC)

So, we have here "ugly" (cryptologist) vs "confusing" (cryptographer). How about cryptanalysts then again ? Rejewski was hardly inventing any codes, but rather "breaking" them.

W, Yup, confusing it is. English has never in its thousand year history as a language distinct from its Germanic/Scandinavian/Latinate roots (as nearly as I can tell) made any rational sense. Attempting to fit Latinate grammar to it is an exercise in driving children mad and the spelling is incoherent, incomprehensible, incredible, and insane. And causes cavities too. As for the 'ugly', it was an attempt to tease out of the back of my mind (on behalf of watching how the language works -- somewhat akin to interrogating one's liver in peculiarity, I think) a reason for discovering that one word is less acceptable to this English speaker than another with equivalent meaning. Not quite a merely aesthetic reaction, but I suppose related somehow (common neural structures?). A similar thing is happening in the back of my brain (or alternatively in Broca's area which isn't quite at the back if I remember correctly) with regard to the definite article when used with NSA. (See Talk:NSA).

Rejewski was certainly a cryptanalyst, but in doing cryptanalysis he was also a cryptographer since he was doing something cryptographic, ie, cryptanalysis. You're right, English is something or other. Be glad Polish makes more sense. (I trust it does?) ww 14:16, 28 Jun 2004 (UTC)

As for Polish usage, historically it was certainly more influenced by German than English, but this trend is getting reversed now. All these words derive from Greek (graphein, analysis, kryptos, logos) anyway.

Today in Polish both 'kryptologia' and 'kryptografia' are in use, the former has a little more "scientific" flavour ("the science of codes"), while the latter is more about their practical exploitation. However with the increasing influence of American English, the word 'kryptografia' is being in much more used now. Surprisingly, a person performing these tasks is called 'kryptolog' and the word 'kryptograf' I think is even incorrect (formally does not exists. But of course peculiarities of Polish usage are irrelevant to English wikipedia. Anyway, in my mind the connotations are following:

• cryptologist = one who researches coding and decoding
• cryptographer = one who codes

W, I've never seen this word used with this limited meaning in English. We use code clerk (for work including cypher work as well, sigh...) or something similarly compound (eg, communications staffer to suggest a particularly awkward one). Looks we have a wider sense for the English cognate of the (nonexistent) Polish kryptograf. Odd how things get passed back and forth (and dropped) between languages, isn't it? ww 13:58, 28 Jun 2004 (UTC)

• cryptanalyst = one who decodes

W, Likewise if the sense you mean is the unraveling of codes as distinct from cyphers. I've always seen it used in English as inclusive of both, never just one. ww 13:58, 28 Jun 2004 (UTC)

In line with these definitions Rejewski was both cryptologist (when he performed some intellectual work) and cryptanalyst (when he was decoding the messages) but never a cryptographer. Hence my doubts. Wojsyl 15:52, 27 Jun 2004 (UTC)

Rejewski himself on terms (in Polish): http://www.spybooks.pl/en/archiwum_tekst.html?id=a6f91165cf09166aa9c20cbc1ada3fb4fbf9 — Matt 00:54, 23 Aug 2004 (UTC)

Single most valuable contribution?

I moved this sentence here for now:

It would not be unreasonable to characterize Rejewski's early 1930s conceptual breakthrough as the single most valuable contribution by an individual to the winning of WWII by the Allies.

Certainly Rejewski's work was highly important, but I think assertions like this are extremely debatable. Of course, if this is a common evaluation of Rejewski's contribution, then it might be good to include a sentence like this, perhaps as a quote from a historian; otherwise I think we'd do well to leave it out. The David Kahn quote certainly helps demonstrate how well-regarded he is by cryptographers. — Matt 04:25, 7 Sep 2004 (UTC)

Some biases

On the principle of parsimony, I regret William Friedman's aging neologism, "cryptanalysis." Why did he need a new term, when there already was a well-established, elegant one, "cryptology" ("the study of secret writing," chiefly for the honorable purpose of "decrypting" or "breaking" others' ciphers or codes)? The proper term then for the creation or use of ciphers or codes was "cryptography" ("secret writing," per se). The two fields are, of course, two sides of the same coin: one studies crypto-systems both to break those of an opponent, and to make one's own resistant to breaking.

Both Anglophones and Polonophones show a marked predilection for conceptual sloppiness. The former like to lump ciphers and codes together as "codes," hence referring to the breaking of either as "codebreaking." Polonophones, doubtless from esthetic considerations of their own, tend to do exactly the opposite: they lump ciphers and codes together as "ciphers," and refer to decryption of either as "deciphering." Such verbal sloppiness inevitably leads to conceptual confusion and should, where possible, be avoided. There is, of course, the third term, "decryption," that may be applied to mixed situations.

Uninitiates, of course, tend to confuse "decipherment" with "decryption"—which is like equating, respectively, a legitimate bank withdrawal with a bank burglary.

But what I perhaps find most objectionable is the imputation of inelegance, stodginess or excessive punctiliousness to any word that contains the Greek root "-logy" or "-logist." Logologist 07:53, 12 Feb 2005 (UTC)

Authoritative account

"the bomba — Polish for "bomb": the most authoritative account, by Cipher Bureau technician Czesław Betlewski, relates..."

For reasons of NPOV, I don't believe that we can assert that this or that account is authoritative or otherwise unless there's some clear consensus in the literature that this is so. My suspicion is that there is no consensus here; there are plenty of other accounts which seem to remain unchallenged: "Bomba is Polish for bomb. Wladyslaw Kozaczuk's book Enigma (University Press of America, 1984, 63) cites a letter from Col. Tadeusz Lisicki, chief of a Polish signal unit, which claims that Jerzy Rozycki named the machine after an ice cream dessert the mathematicians were eating at the time. The bomba dessert was a round ball of ice cream covered in chocolate and resembled an old-fashioned bomb. However, in an article Rejewski himself says, "For lack of a better name we called them bombs." ("How Polish Mathematicians Deciphered the Enigma," Annals of the History of Computing, v.3, n.3 (July 1981): 226.) Finally, a U.S. Army document describing the Polish Bombes claims, "When a possible solution was reached a part would fall off the machine onto the floor with a loud noise. Hence the name 'bombe'." (6812th,10.)" [1] (http://www.nsa.gov/publications/publi00016.cfm). — Matt Crypto 08:29, 12 Feb 2005 (UTC)

The phrasing as it now stands, without "most authoritative," is fine.

It's questionable whether Tadeusz Lisicki had factual grounds for his assertion: he attributes the device's naming to Jerzy Różycki, whom he never met. (Różycki died in the Mediterranean in 1942.) Lisicki was never associated with the Cipher Bureau. Betlewski was. Barring convincing documentation, how the Americans derive the name is irrelevant, since they hadn't named the device in the first place.

I doubt that Rejewski or Zygalski would have discussed Enigma decryption, much less the naming of the bomb, with Lisicki during the war, having been sworn to secrecy by their employers. Lisicki received information from Rejewski after Enigma decryption had become public knowledge and Lisicki offered to advocate for the Poles' primacy. Logologist 09:10, 12 Feb 2005 (UTC)

Interesting. It'd be good to add Betlweski's account to the various theories listed at Bomba (cryptography) — would you be able to provide a reference for this? — Matt Crypto 09:53, 12 Feb 2005 (UTC)

Betlewski's version appears in Kozaczuk's 1984 Enigma, p. 63, note 1: "B.S.-4 workers also referred to the device, irreverently, as a 'mangle' ([Polish:] magiel) or 'washing machine' (pralka), on account of the characteristic muffled noise that it produced. (Oral account by Czesław Betlewski.)"

Betlewski also appears on pp. 212-15, in relation to security considerations under German occupation and drastic measures employed to protect the Enigma-decryption secret.

Incidentally, a diagram of a Polish cryptological bomb appears on p. 289 which I believe has also figured in other publications; an adaptation might serve as a helpful illustration to the "Bomba" article. Logologist 05:14, 13 Feb 2005 (UTC)

Thanks for the reference. Would you be able to also provide the source s for the Rejewski, Good and Welchman quotes? I've recently realised that it's quite a good idea to provide sources for quotes and the like (even though it's quite tedious!). Yes, the bomba picture (if it's the same as the ones I've seen) would be quite useful if someone could create one for us. — Matt Crypto 22:55, 13 Feb 2005 (UTC)

Sources provided.

Thanks! — Matt Crypto 10:23, 14 Feb 2005 (UTC)

The drawings of Rejewski's cyclometer and bomb that appear in Kozaczuk 1984 (pp. 284, 289) are reproduced identically in Jan Bury's "The Enigma Code Breach," which you've linked to "Marian Rejewski." They have, in fact, appeared in so many places that they probably may be considered public-domain.

While we probably can't consider them public domain (unless the author has specifically indicated as such) their widespread use does speak in favour of a Fair Use case (to some extent). I think it would be preferable to create our own, though. — Matt Crypto 10:23, 14 Feb 2005 (UTC)

Bury's piece, though generally accurate, is occasionally misleading (e.g. the cyclometer came long before the bomb) and of course requires translation from "Poglish" to English (e.g. "auxiliary connectors' plate at the front panel" = "plugboard"). Logologist 06:35, 14 Feb 2005 (UTC)

Extended Rejewski quote

I'm not sure that the extended Rejewski quote is the best way here. Short quotes are clearly very useful, but the job of an encyclopedia article is to summarise primary sources; here, we seem to be incorporating Rejewski's narrative to tell the story for us. I don't believe the events are so contentious that we can't synthesise our own summary of them. — Matt Crypto 10:23, 14 Feb 2005 (UTC)

There has been an awful lot of confusion and speculation (for over 30 years!) about just what it was that Rejewski did. I thought that hearing about it straight from the horse's mouth might help clear things up. But perhaps someone could attempt a deft (and accurate) summary — and perhaps we might consider transferring the extended quotes to the WikiQuotes, for convenient reference? Logologist 12:27, 14 Feb 2005 (UTC)

Yes, sounds like a good plan to me, and Rejewski's account does seem the obvious starting point for a summary of the events! — Matt Crypto 12:17, 15 Feb 2005 (UTC)