81

I've seen/read numerous old telegraph messages. They contain a lot of spelled-out punctuation. Why didn't they simply have the most basic few characters as part of the code, or at least turn a "STOP" into "." and "QUOTE" into a '"' on the other end?

This might make it harder to read. The reason seems unlikely to be technical, because if they can transmit 26 letters as Morse code, they can also transmit more characters OR use special "code tags" formed by the existing ones. For example, the word "TCC1" could be short for "Telegraph Character Code 1", meaning a period, which is never displayed as a "STOP" in the final message.

Even if the reason was that they couldn't agree on a standard, it would still seem better to get the actual code "TCC1" printed out rather than:

WE TRIED TO MAKE HIM STOP STOP AND WE COULD NOT FIND A WAY TO STOP HIM STOP STOP AT ONCE AND COME HOME STOP

As opposed to:

WE TRIED TO MAKE HIM STOP TCC1 AND WE COULD NOT FIND A WAY TO STOP HIM TCC1 STOP AT ONCE AND COME HOME TCC1

Or (for telegraph printers supporting my theoretical standard):

WE TRIED TO MAKE HIM STOP. AND WE COULD NOT FIND A WAY TO STOP HIM. STOP AT ONCE AND COME HOME.

Why did they not adopt such a scheme?

Michael MacAskill
  • 105
  • 3
Jolin
  • 651
  • 1
  • 5
  • 4
  • 14
    The Morse code (which is what you are complaining about) was invented about half a century before "machines that could remotely transmit messages across the world and print them out on the other end" were invented. – kimchi lover Feb 13 '20 at 02:11
  • 1
    @kimchilover: The fax machine was patented in 1843 by Alexander Bain as the "Electric Printing Telegraph". – Pieter Geerkens Feb 13 '20 at 02:50
  • 24
    To clarify for the misguided downvoters: Morse code has included codes for punctuation marks since its early days in the mid-1800s. The original code for a period was six shorts. This was changed to three groups of two shorts each around 1900 and to the modern iambic trimeter in the 1930s. See [here][1], referencing Dots and Dashes, vol. 35 nbr 3 (Summer 2009). While this isn't strictly an answer to the question, it addresses a related point about which a commenter and many of his upvoters seem ignorant, and doesn't fit in a comment. – C Monsour Feb 13 '20 at 02:54
  • 2
    Since the moderator who moved my answer to a comment didn't know how to copy the link to the reference, here it is: http://www.morsetelegraphclub.org/wirechief/ – C Monsour Feb 13 '20 at 12:12
  • 2
    I remember hearing somewhere (I can't source this so not making an answer) that telegraph operators used to charge by the sentence. By using "STOP" instead of a period, you could cram several sentences together and pretend they were one, and get a reduced rate. – Darrel Hoffman Feb 13 '20 at 19:02
  • @CMonsour: Would telegraph operators have sent a "stop" as a punctuator with normal letter spacing to either side, or as simply an extra-long space, or would they combine the punctuation with extra space? A newfangled .-.-.- period would take 20 units of time to send, while a word space would take 7, so a space 2.5 times as long as normal would seem like it would be faster to send than a period and more visually distinctive on a marking tape. – supercat Feb 13 '20 at 22:02
  • 1
    It seems obvious to me that if you're transcribing as you go, you don't know you've gotten STOP until you get the P, so you can't retroactively change that to a punctuation mark. Every time they got an S, they would have to wait and see if the next character was a T, if it wasn't then they could write the two letters down, and so on - seems error-prone. – Wossname Feb 14 '20 at 00:10
  • Counterexample: Victor Hugo, or these: https://quoteinvestigator.com/2014/06/14/exclamation/ – Rob Feb 14 '20 at 02:55
  • 3
    E Blackadder: "MR C CHAPLIN SENNETT STUDIOS HOLLYWOOD CALIFORNIA CONGRATS STOP HAVE DISCOVERED ONLY PERSON IN WORLD LESS FUNNY THAN YOU STOP NAME BALDRICK STOP YOURS E BLACKADDER STOP P S PLEASE PLEASE PLEASE STOP" C Chaplin: "TWICE NIGHTLY SCREENING OF MY FILMS IN TRENCHES EXCELLENT IDEA STOP BUT MUST INSIST E BLACKADDER BE PROJECTIONIST P S DONT LET HIM EVER STOP" – CJ Dennis Feb 14 '20 at 09:54
  • 1
    @Wossname: What do you mean "get to the P"? I would expect that telegraph operators would be able to see the entire written word, or hear the entire word spoken, before they start sending any of it, and I don't think STOP is sent as "... - --- .--." unless maybe in a context like "I DECIDED TO STOP AT THE NEAREST TRUCK STOP STOP", in which case the first two would be sent as words but the third via other means. – supercat Feb 14 '20 at 16:29
  • @supercat This would apply to the person receiving the message, not the one sending it. As the person on the receiving end would not have paper to look at or a voice to listen to, they can't "look ahead" as it were. – Darrel Hoffman Feb 15 '20 at 15:41
  • @DarrelHoffman: If STOP is sent as .-.-.- instead of ... - --- .--., how could the recipient wait for the letter "P", since it would never be sent? – supercat Feb 15 '20 at 19:12
  • @supercat Which is exactly the point that Wossname was making. The recipient wouldn't be able to wait since they wouldn't know it was a STOP until they received a 'P'. (Assuming they don't use punctuation for whatever reason.) – Darrel Hoffman Feb 17 '20 at 05:48
  • @DarrelHoffman: Why would the transmitter send the letters S-T-O-P? – supercat Feb 17 '20 at 14:47
  • In all caps, that first block quote is actually the easiest to read. Punctuation without upper and lower case is easily missed. And using words that don't exist that no one's agreed on helps nobody. Are they trying to call me a TOOL but it was lost in translation? Maybe people stopped using the word stop so much when they knew that the world stopped on the word stop for stops, full stop – Mazura Feb 29 '20 at 05:03

3 Answers3

86

A 1928 booklet on HOW TO WRITE TELEGRAMS PROPERLY has this to say concerning the use of STOP (emphasis mine):

If it seems impossible to convey your meaning clearly without the use of punctuation, use may be made of the celebrated word "stop," which is known the world over as the official telegraphic or cable word for "period."

This word "stop" may have perplexed you the first time you encountered it in a message. Use of this word in telegraphic communications was greatly increased during the World War, when the Government employed it widely as a precaution against having messages garbled or misunderstood, as a result of the misplacement or emission of the tiny dot or period.

Officials felt that the vital orders of the Government must be definite and clear cut, and they therefore used not only the word "stop," to indicate a period, but also adopted the practice of spelling out "comma," "colon," and "semi-colon." The word "query" often was used to indicate a question mark. Of all these, however, "stop" has come into most widespread use, and vaudeville artists and columnists have employed it with humorous effect, certain that the public would understand the allusion in connection with telegrams. It is interesting to note, too, that although the word is obviously English it has come into general use In all languages that are used in telegraphing or cabling.

"Stop" is of course never necessary at the end of a message.

So the goal was clarity of the message.

Since there seems to still be some question about why the actual word STOP is being used, we can look at the Wikipedia article on Full Stop for more information:

The word period was used as a name for what printers often called the "full point" or the punctuation mark that was a dot on the baseline and used in several situations. The phrase full stop was only used to refer to the punctuation mark when it was used to terminate a sentence.

Again, clarity. A period had several uses, but the Full Stop is the proper term for the item terminating a sentence. No errors or misunderstandings, STOP meant the end of the sentence.

justCal
  • 39,636
  • 2
  • 116
  • 170
  • 21
    That explains why there had to be a delimiter, but not why the delimiter had to be a common English word. – Sneftel Feb 13 '20 at 11:01
  • 4
    @Sneftel A period is, in British English, called a "full stop". – Williham Totland Feb 13 '20 at 11:04
  • 6
    @WillihamTotland Yes, I'm aware of that. It doesn't really speak to my point. – Sneftel Feb 13 '20 at 11:09
  • 6
    @Sneftel Except it does: The delimiter is the word 'STOP' because the word 'STOP' is the name for the delimiter being spelled out. It's the same reason ',' is replaced by 'COMMA' and ':' by 'COLON': Those are the names of the things being replaced. – Williham Totland Feb 13 '20 at 11:10
  • 17
    @WillihamTotland I'll try to explain again. The quote explains why it was useful to adopt an artificial convention in which a full stop was rendered as a sequence of characters rather than as a ".". But that sequence of characters could have been anything. Why choose a word which (among its many other uses) literally describes the replaced punctuation, rather than an invented sequence which would unambiguously describe it? – Sneftel Feb 13 '20 at 11:14
  • 15
    ""Stop" is of course never necessary at the end of a message." -- ohh, that just sound like an invitation for truncated messages. Unless they had some surrounding framing to detect that, of course. – ilkkachu Feb 13 '20 at 12:48
  • 9
    @Sneftel Well, it's a lot easier to garble "TCC1" into "TCC2" than it is to garble STOP into COMMA, so that's a plausible argument (but realistically, they probably didn't think of machine codes as a thing at that point). – user3067860 Feb 13 '20 at 14:33
  • 7
    @ilkkachu They did. There actually is a protocol for sending messages via telegraph through a network, and for moving from one telegram to another telegram, and checking if the number of received words was correct and so on. Additionally, after the end of a message, you'd actually also have a signature field, with the name of the person who sent the message, so it would be easy to notice the truncation. – AndrejaKo Feb 13 '20 at 15:12
  • 8
    Sure, but it's a lot harder to confuse QXR with JWF -- or with any English word. – Sneftel Feb 13 '20 at 15:13
  • 2
    But it's easier to understand telegrams that are read out aloud, when you use "STOP" and "QUERY" instead of "QXR" and "JWF". And quicker to send them. – Rupe Feb 13 '20 at 16:20
  • 25
    @Sneftel one advantage STOP enjoyed over your suggestions is that it didn't need parties to agree what the code would be. Properly crafted, a telegram containing STOP would be understood by a recipient who had been party to no such arrangement, simply because the full stop it represented would be the most straightforward interpretation of the English word employed at that point in the text. – Will Feb 13 '20 at 18:38
  • @will makes a great point. – Ben Feb 13 '20 at 21:00
  • 2
    It's a bit ironic that someone discussing clear communication wrote "emission" rather than "omission". – Acccumulation Feb 14 '20 at 04:50
  • 1
    @Sneftel you "intentionally" overlooked the first paragraph in the answer - punctuation was generally frowned upon in telegraph use - so they used words to express the limited punctuation – eagle275 Feb 14 '20 at 12:23
  • @eagle275 Yes, that was the situation I and the OP were talking about: why (real) words were seen as the best substitute for punctuation. – Sneftel Feb 14 '20 at 12:27
  • 2
    @Sneftel It seems obvious to us now, looking back almost 200 years, that delimiters chould have a separate or arbitrary code to prevent confusion, but is there any precedent for that? Just because an idea is good or obvious doesn't mean that anyone actually thought of it. Even if someone did, there are a lot of social/cultural reasons a particular technical solution may not have caught on, least of all being that learning to read a message and figure out which "STOP"s were words and which were punctuation is just easier than learning several more unique codes. – nexus_2006 Feb 14 '20 at 15:45
  • 2
    @nexus_2006 There's actually significant precedent, in the form of naval flag signalling, flag semaphore, and shutter semaphore. The electric telegraph replaced already-mature telecommunications systems. See, as a famous example, https://en.wikipedia.org/wiki/England_expects_that_every_man_will_do_his_duty . – Sneftel Feb 14 '20 at 17:09
  • I don't know a lot about those signalling systems, but after a quick perusal of naval flag and semaphone wiki pages, I don't see anything about punctuation or "message finished" signals, except a "pause/space" signal in flag semaphore. I looks to me like telegraph sent much longer and complex messages than either of these systems. The "STOP", "COMMA", "SEMICOLON", etc. codes seem to be an innovation favoring human readability and not having to look up a lot of arbitrary codes in a code book. – nexus_2006 Feb 16 '20 at 13:34
69

Because telegraph was a manually routed transmission. Like the original Ethernet of the 1980's, one had long cable runs with stations tapping the cable along its length. Transmission to another station on the same cable run was single hop, but to a station on another cable run would be multi-hop.

Unlike Ethernet however, with it's automatic repeaters, telegraph had manual repeaters who would receive a message on one cable and retransmit it on another. For a Trans-Atlantic transmision this might involve several hops on each side of the trans-Atlantic cable:

  • Evanston to Chicago

  • Chicago to New York

  • New York to Belfast

  • Belfast to London

  • London to Paris

  • Paris to Rheims

  • Rheims to Verdun

In all these retransmits, there was a very real danger that single punctuation marks might get missed, thus garbling the message. By introducing distinct words such as STOP, COMMA, SEMICOLON for the punctuation marks a degree of redundancy - essentially error-checking - was introduced that made such errors vastly less likely.

That it was the military that first made this practice standard is not surprising. When giving orders, competent commanders go to great lengths to ensure that the orders are direct and unambiguous. An example of the consequences of even a simple failure, two orders arriving out of sequence, is well known from the first days of the 1809 campaign in Bavaria.

Napoleon sent a first message to Berthier from Paris by semaphore, which was delayed for more than a day by cloudy conditions near Strasbourg. Then he sent a second, more detailed, message by courier which arrived to Berthier first. As a consequence of the orders arriving out of order (but not being recognized as such), Berthier took the more general instructions as an amendment of the detail rather than the other way around - resulting in Davout's corps remaining at Regensburg two days longer than intended by Napoleon.

The ensuing correspondence between Berther and Davout is well described in Volume 1 of John H Gill's Thunder on the Danube, as the two marshals attempt to sort out, long distance, the true intent of Napoleon's orders.

Why only puncutation you might ask? Because normal language already contains a great deal of redundancy, both in spelling and grammar, as evidenced here:

It deosn't mttaer in what oredr the ltteers in a word are, the iprmoatnt tihng is that the fisrt and last ltteer be in the rghit pclae. And I awlyas tghuhot slpeling was ipmorantt

Finally - why were English words used instead of special code? Because the sender was charged for the message by the "word" - and every character of a "special code" was it's own word. And why was that you ask - because words can be processed faster and more accurately than special codes. Words, as noted above, contain error-checking redundancy that special codes cannot. The difficulty is not at the sending end so much as at the receiving end, where the receiver cannot utilize any obvious redundancy to ensure accuracy.

The consequential combination of both lower cost and improved accuracy (the whole point of replacing single character punctuation after all) meant it was never going to be useful, in the large, to use special codes rather than words.

Further examples of telegraphese and commercial (telegraph) code aimed at both decreasing cost and improving clarity and accuracy of telegraphed messages..

Pieter Geerkens
  • 72,560
  • 9
  • 210
  • 330
  • 1
    Comments are not for extended discussion; this conversation has been moved to chat. – T.E.D. Feb 13 '20 at 15:23
  • 7
    This answer seems to fall prey to the same issue encountered with the other answer written by justCal-- it explains why a character sequence instead of a literal '.' was needed, but it does not explain why they used an English word 'STOP' instead of a codephrase for increased readability, such as 'TCC1'. – Onyz Feb 13 '20 at 19:21
  • 7
    @Onyz: Addressed. I forget that many are not old enough to understand how telegraph worked - this point seemed so obvious to me as to hardly even be worth mentioning. – Pieter Geerkens Feb 13 '20 at 20:21
  • 4
    That jumbling hing is not true anyway. The condition for first+last letters reduces the jumbling considerably due to the short words. And all used examples are only moderately jumbled anyway. Order the letters alphabetically and/or change syllable structure (e.g. by moving the vowels together), and longer words became very hard. ipmorantt is much easier to read than iaotmnprt. The point the answer is still correct, an error in a telegram will most likely only affect neighboring characters. And special codes require more mental work to recognize. – Chieron Feb 14 '20 at 07:49
  • Either I remember it wrong or I was told once that people were charged per sentence for telegraphs, so they made one long sentence by leaving out all punctuation marks. But of course, now that I think about it again, that trick wouldn't have worked for long enough to still be a meme a century later. – Fabian Röling Feb 15 '20 at 01:21
2

Telegrams were charged per word, and the "stop" and other punctuation marks were counted as a word. Spelling them out was a way to communicate that this was important enough to be charged for. It prevented arguments and inaccuracies of how much a telegram cost, and also prevented people from cheating their system by developing a code whereby combinations of punctuation marks would convey a meaning.

I am a guest 38303143
  • 29
  • 1
  • 4
    "and also prevented people from cheating their system by developing a code whereby combinations of punctuation marks would convey a meaning" - Not really, you can just as easily "cheat" by using a code where combinations of normal letters convey a meaning (aside from their normal meaning). Like "ADD SEVEN" could mean "Attack at Dawn, Seven days from now", if both you and your recipient have agreed to that code. – aroth Feb 14 '20 at 01:18
  • 7
    All special characters, including the original punctuation marks, were a word and charged as such. There was no cost change by substituting "STOP" for ".". – Pieter Geerkens Feb 14 '20 at 14:48
  • 5
    At least in the world of jokes, you could also sign your name to a telegram for free. "Oh, no message," says the college freshman; "just sign it 'Hellopop Pleasesendmoney.'" – Quuxplusone Feb 14 '20 at 17:03
  • 3
    If only there was some widely known method of encoding text in punctuation such as dots and dashes, eh? – Chronocidal Feb 28 '20 at 17:09
  • 2
    @Chronocidal STOP STOP SPACE SPACE STOP HYPHEN SPACE HYPHEN HYPHEN STOP SPACE STOP HYPHEN STOP SPACE STOP SPACE STOP SPACE SPACE STOP STOP STOP SPACE HYPHEN SPACE HYPHEN HYPHEN HYPHEN SPACE STOP HYPHEN HYPHEN STOP – user253751 Mar 02 '20 at 15:25