Sunday February 8, 2004

"Omnilingual", by H. Beam Piper

[First in a series about Linguistics in Science Fiction. Fair warning: I plan to spoil the ending.]

How would you decipher texts in an unknown language, written in an unknown writing system? H. Beam Piper's short story "Omnilingual", originally published in 1957, is about an archaeological expedition on Mars, exploring the remains of a dead civilization. The expedition's linguist is confronted with a seemingly insurmountable problem: texts in an ancient language with no remaining speakers, and for which no bilingual text exists. What's an Earth linguist on Mars to do?

This story is good-old-fashioned science fiction. Mars is almost, but not quite, dead: the Martian civilization is gone and the Martians are extinct, but there's still some oxygen in the atmosphere, and some life survives in the dry ocean basins. The expedition is large and well equipped, and a bit reckless—for working space, they're in the habit of caulking up all the cracks in 50,000-year-old buildings, pressurizing them, and then smoking cigarettes in the resulting pure-oxygen environment, presumably all the while laughing in the face of danger.

Our heroine is Martha Dane, Archaeologist. She's been collecting scraps of Martian writing, trying to decipher the language, but it doesn't look promising.

"There is no Rosetta Stone, not anywhere on Mars. A whole race, a whole species, died while the first Cro-Magnon cave-artist was daubing pictures of reindeer and bison, and across fifty thousand years and fifty million miles there was no bridge of understanding."

The few written materials that Martha has found include what appears to be a magazine (on some kind of long-lasting silicone rather than on paper, Piper is careful to mention). She's pored over it (while smoking!) and figured out the basics of the writing system:

It was readable, in the sense that she had set up a purely arbitrary but consistently pronounceable system of phonetic values for the letters. The long vertical symbols were vowels. There were only ten of them; not too many, allowing separate characters for long and short sounds. [A bit of a leap, I think—maybe they have rounded and unrounded versions of five vowels.] There were twenty of the short horizontal letters, which meant that sounds like -ng or -ch or -sh were singles letters. [A strange way to phrase it—I think Piper means every consonant is written with a single letter, because how could she know if those sounds existed in Martian?] The odds were millions to one against her system being anything like the original sound of the language, but she had listed several thousand Martian words, and she could pronounce all of them.

The magazine's title, in her scheme for pronouncing the writing system, is Mastharnorvod Tadavas Sornhulva. But what do the words mean? She's found some illustrations with captions, but there's no obvious correlation between the captions and the contents of the pictures. Worse, another character points out that captions probably aren't going to help much:

"A caption is intended to explain the picture, not the picture to explain the caption. Suppose some alien to our culture found a picture of a man with a white beard and mustache sawing a billet from a log. He would think the caption meant, "Man Sawing Wood." How would he know that it was really 'Wilhelm II in Exile at Doorn?'"

Martha thinks she has identified the (conveniently decimal) numerals and, if she's made a correct guess that part of the magazine's cover includes the date, the name of a Martian month, but other than that, she's stuck. Without a true bilingual text like the Rosetta Stone, or even proper names correlate in the undeciphered texts, how can she hope to go further?

Things start to go her way when the expedition cracks open a building that appears to be a university. A huge store of books is found (the university's library). This gives her much more data to play with, but it's still not the key Martha needs. [Aside: Piper's writing in the 50's, so this won't have occurred to him, but I wonder what kind of results you could get by throwing a huge wad of cheap early-21st-century computing power at a corpus of this size.]

Her break comes when they find the department of physics/chemistry (Martians don't distinguish the two). On one of the walls, they find a diagram of a simple atom and a table of words and numbers:

Ninety-two! That was it; there were ninety-two items in the table on the left wall! Hydrogen was Number One, she knew; One, Sarafaldsorn. Helium was Two; that was Tirfaldsorn. She couldn't remember which element came next, but in Martian it was Sarfalddavas. Sorn must mean matter, or substance, then. And davas; she was trying to think of what it could be.

Martha's companions are able to confirm it's really the Periodic Table by pointing out that the values for the atomic weights are correct. With their help, she decides davas must be the word for "metal". This really is a 50's SF story, then, because it has turned out that Science Is The Key!

"That isn't just the Martian table of elements; that's the table of elements. It's the only one there is...Look, hydrogen has one proton and one electron. If it had more of either, it wouldn't be hydrogen, it'd be something else. And the same with all the rest of hte elements. And hydrogen on Mars is the same as hydrogen on Terra, or on Alpha Centauri, or in the next galaxy—"

Piper claims that the Periodic Table is better than a bilingual text—it's the same everywhere, so it's "omnilingual". I think he may be overestimating the extent to which such a find would be the key to figuring out an entirely unknown language (how are they going to figure out the non-technical vocabulary?), but it's a clever twist and makes for a fun story. The details of the decipherment problem are pretty plausible too, although Piper stacks the deck in his characters' favor—most notably, he makes Martians basically physically identical to humans, so Martha has a starting point to make some guesses about the phoneme inventory. Piper was inspired, I think, by way that real historical decipherments were based on finding the narrowest of toeholds, and climbing on from there: the names of pharaohs for the Rosetta Stone, or the names of Minoan cities for Linear B. Piper's insight is that the laws of science can serve the same purpose. It's an interesting point about linguistic universals, before they became such a hot topic in linguistics: all languages, even Martian languages, are spoken by people whose bodies and experiences have to conform to universal physical laws.

I am The Tensor, and I approve this post.
03:52 AM in Linguistics in SF | Submit: | Links:


TrackBack URL for this entry:

Listed below are links to weblogs that reference "Omnilingual", by H. Beam Piper:

A new blog, apparently language-oriented, called Tenser, said the Tensor is beginning a series about linguistics in science fiction with a post on a good H. Beam Piper story:How would you decipher texts in an unknown language, written in an... [Read More]

Tracked on Feb 16, 2004 3:10:57 PM

A new blog, apparently language-oriented, called Tenser, said the Tensor is beginning a series about linguistics in science fiction with a post on a good H. Beam Piper story:How would you decipher texts in an unknown language, written in an... [Read More]

Tracked on Oct 27, 2009 12:29:12 PM


For more info on Piper, try my URL.
The reason the Martians are human is that in Piper's set of stories the earth was settled from Mars 70,000 years ago (it tied in with his Paratime series.)
The idea is fairly widespread in SF -- the point is made in an episode of Stargate -- but actually not that well understood in the academy. Back in the 80s Carl Sagan attended a conference in (then) Soviet Georgia where a Marxist British historian actually asked "What if hydrogen isn't hydrogen in another Galaxy?" Crick (of Watson and Crick fame) totally blew up; he would have done better to cite Piper.

Posted by: John Costello at Feb 16, 2004 3:45:22 PM

A nicely written synopsis. An intriguing story and concept.

This reminds me of a book by Eleanor Arnason: A Woman of the Iron People. In this story the problem becomes understanding languages and cultures that are still alive (and technically underdeveloped).

Posted by: keto at Feb 16, 2004 6:44:55 PM

There's a great new astrobiology blog, run by newspaper editor Rob Bignell, at It includes roundups of the latest news from the various scientific fields that form astrobiology and information about SETI.

Posted by: Kyle Janison at Jul 5, 2005 5:24:18 PM

I read the story for the first time about a year ago. But for all my enjoyment of the story and it's solution one little thing about it kept sticking for me...How did Martha manage to TRANSLITERATE Martian in the first place.

Posted by: Bill Morse at Dec 6, 2005 5:36:23 PM

How did Martha manage to TRANSLITERATE Martian in the first place.

Maybe she cheated? That is, maybe she computed the frequencies of the vowels & consonants, and assigned them values based on English letter frequencies?

Another possible solution to the Martian language problem as a whole: She could have looked for children's books. After all, such books would be *designed* to instruct those who are unfamiliar with the written language in the language - so if there's a picture of a man sawing wood, it most likely wouldn't be captioned with an abstract historical reference (that a child would have no way of knowing), but would have words referencing the nouns and the verbs being depicted.

Now I want to see "One Thoat, Two Thoat, Blue Thoat, Green Thoat", by Dr. Foiff.

She should also have looked for linguistics texts. Somewhere in there would be a book with pictures of a mouth pronouncing the letters, so she would have a much better idea of what the phonemic values actually were.

And finally, while the periodic table is indeed universal, all that would get her for certain is the numbers (chemical weights) - which she already had anyway. Piper cheated and made the names of the elements extremely logical, connecting them to the type of substance they were. In our world, the element names are pretty random.

See also this:

and this (which are names of compounds rather than elements, but nevertheless):

Posted by: Owlmirror at Mar 27, 2006 9:54:42 PM

Useful . . . good. This article helps.

I teach SF here, at a university in Serbia, Europe . . . There will be an article, I trust, on Jack Vance’s “Languages of Pao”?

Posted by: Prof. Dr. Alexander B. Nedelkovic at May 19, 2006 1:21:49 PM

Already written. Look in the sidebar to the right, under "Linguistics in SF".

Posted by: The Tensor at May 19, 2006 3:40:13 PM

Thank you. I got it. A nice article too.

Posted by: Prof. Dr. Alexander B. Nedelkovic at May 20, 2006 1:44:51 PM

Omnilingual is now online at Project Gutenberg!

Posted by: Steve at Oct 5, 2006 9:26:06 PM

For the record, I have prepared an edited version of the story removing some of its 1950s-isms. It's available at .

Posted by: John Cowan at Oct 22, 2009 4:31:02 PM

I cant believe this ever got printed. First - what "marcian language"?! Seriously? The entire planet spoke the same language? Aha....
Second - oh please, they found entire Library! If all they would have found was that periodic table, then yes it would be a little more exciting to see it's being worked through. But they found entire university's library!! What, there no anatomy books with pictures and clear name of parts/organs for it? There no math books? (sure, the numbers and letters are unkown but the system is still the same. i mean Xsquare + Ysquare = Zsquare next to triangle picture will mean the same if you replace letters with flowers and numbers with bugs).
Yeah, so i challenge you to learn "Earth language" by using periodic table. Good luck.
p.s. if all his books are anything like this story, i think i will skip on the rest of them.

Posted by: Svant at May 26, 2010 7:52:14 AM