Most recent comments
Moldejazz 2018
Camilla, 11 months, 1 week
Romjulen 2018
Camilla, 1 year, 6 months
Liveblogg nyttårsaften 2017
Tor, 2 years, 6 months
Jogging og blogging
Are, 3 years, 5 months
Liveblogg nyttårsaften 2016
Are, 3 years, 5 months
Reading in dark times
Are, 3 years, 7 months
Moldejazz 2016
Camilla, 3 years, 11 months
Karoline, 3 years, 11 months
Melody Gardot
Camilla, 11 months, 3 weeks
Den årlige påske-kommentaren
Tor, 1 year, 2 months
50 book challenge
Camilla, 6 months, 1 week
Ten years ago
Løp og bli

Linguistic old wives’ tales

Since it seems the boys from Calcutta like reading linguistics-related articles, I thought I'd post this here. A couple of days ago a friend asked sent me a message asking for my opinion on something a guy had said on a message board concerning moccasins. This word is well known as an example of a word that English borrowed from a Native American language. The claim was that the Native Americans themselves had borrowed it from Scottish Gaelic.

Interestingly the gaelic expression for "my shoes" is mo casain (or something like that) which is pronounced almost exactly like the red indian "moccassin", suggesting that the earliest lingual exchanges between our cultures were between the gaels and the red injuns.

Another poster later "explained" how this came about.

During a Gaidhlig clas today, we did a letter from Leabhar nan Litricheacn (litir 90). Basically it explained very well how this came about (Mocassin). Plenty of Indian tribes only contact with whites would have been with Gaels (As in other areas of North America only with the French or English speakers), there are plenty of cases where Indians also learned and spoke Gaidhlig and sackloads of cases of intermarrying. The letter explains that an Englishman who was something of an anthropoligist came accross some Indians whose only previous contact had been with Gaidhlig speakers, the Englishman on noticing the Indians brightly coloured shoes pointed to his feet and asked what they were called, and got the Gaidhlig answer "mo chasan" back because the Gaidhlig speaking Indian assumed the Englishmans tongue was just a funny form of Gaidhlig (white mans language) that hed never heard before which kinda makes sense. To this day the Mic Mac people (Red Indian) of Nova Scotia speak Gaidhlig.
Co aig a fios, who knows.

Someone else called bullshit:

There are a number of problems with this, Neilly.

1. As mentioned before - dates. You don't give a date for this letter you mention, but, as has already pointed out on this thread, the first appearance of 'moccasin' (as 'Mekezin') in English is 1609, in an English translation of a French work: Nova Francia. We're dealing with a very tight time-frame for the word to move from Gaelic to a number of Native American languages (the form we have comes from Powhatan but closely related words are known in Narragansett, Micmac and Massachusett) and to replace whatever term those languages already had for 'shoe'. Also, I'm unsure how you can state that "Plenty of Indian tribes only contact with whites would have been with Gaels." Cognates of 'Mocassin' are know from related Native languages across large parts of North America (of which more below).

2. Any story that includes phrases like: "The Englishman asked what they were called ..." has got to be treated with caution (see, for example, folk etymologies for 'Canada' and 'Kangaroo'). A statement like: "because the Gaidhlig speaking Indian assumed the Englishmans tongue was just a funny form of Gaidhlig" begs the question, 'How can we possibly know what the Gaidhlig-speaking Indian assumed?' Anyway, as noted above, the word came into English via French - see the OED.

3. In the Massachusett language, 'Mokussinash' was the word for 'shoes', and the singular form was 'Mokus' ('shoe'). It's vanishingly unlikely that a loan phrase would be slotted into the native grammar structures in that way - and the singular form is moving still further away from the phonological coincidence of 'mo chassan'.

As I've said before, I'm not 'dissing' Gaidhlig, I'm just pointing out that etymology is a science (albeit an inexact one), and romantic notions can't go unchallenged and uncorrected - that would be patronizing in the extreme. ;)

One of the posters then asked me for an opinion. This is what I wrote:

Hi John,

I don't know a lot about the etymology of that particular word, but I do know how to tell good linguistic arguments from bad.

It's always tempting to jump on a similarity between two words (or short phrases) in two different languages which have similar meanings, and proclaim that you've "proved" that they're related or that there was contact between speakers much earlier than anybody knew. These things are always worth checking out – we'd never discover anything if people didn't follow up hunches. But in order to convince the academic world, you need a lot more than that – you can't just sit back and rewrite history on the basis of three similar-sounding syllables.

As I understand it, the claim here is for a single borrowing, rather than a genetic relationship (i.e. Powhatan and Gaelic descended from the same ancestor language – a claim which would require a lot of evidence!) This at least reduces the standard of proof required somewhat. But just finding a word that sounds similar could, as has apparently been pointed out, just be a coincidence. This is the "null hypothesis" – the default assumption which must be assumed in the absence of sufficient evidence for any of the others. Let's think about how else we could explain the resemblance. What does the claim involve?

"Moccasin" has, as was pointed out, cognates all across Eastern Algonquian – it's not a word that's pecular to Powhatan. This means either:

(a) contact a very long time ago between Gaelic speakers and speakers of the proto-language which eventually became Powhatan (spoken in Virginia), Massachusett (spoken, surprise surprise, in Massachusetts – 500 miles away from the Powhatans), and Micmac (southern boundary Massachusetts but extending hundreds of miles north of there);

(b) Gaelic-speakers in North America more recently, but with a very wide distribution and enough people in each of the relevant places to have an effect on the language;

(c) word introduced in Virginia (where documented in 1609, it seems) and spread like absolute wildfire throughout the widely scatted Algonquian tribes, eventually ending up being used everywhere between Nova Scotia and Virginia (about 1000 miles overland);

or (d) it's just a coincidence. Let's examine the likelihood of each of these in turn.

(a) Proto-Algonquian is thought to have been spoken 2,500-3,000 years ago. At this time, Proto-Celtic was (probably) just starting to emerge as a separate language. Europe was at this time in the Iron Age, and certainly in no shape to be sailing across the Atlantic. If they went overland, they'd need to walk all the way across Siberia, over the Bering Bridge to Alaska and all the way to the opposite side of North America without leaving any traces of their presence along the way. I think this idea can safely be dismissed as fantasy.

(b) Thousands of Gaels stretching 1000 miles from Virgina to Nova Scotia? We would know about this. It would require half the population of the Highlands and Islands to have gone across with the very first colonists. If this happened, why aren't there 100 Gaelic words in the Algonquian languages instead of just this one? And why isn't there a sizeable Gaelic-speaking community in Virginia today? Why, for that matter, isn't Gaelic the official language of the USA?

Lots of Gaels did, of course, move to Nova Scotia, but that was about 150 years after the word "moccasin" is recorded in Powhatan.

(c) The most obvious reason for this to happen would be if the Algonquians didn't have shoes until they first met a white person, who happend to be a Gael. You can imagine the exchange: "What are those on your feet, paleface?" "My shoes" (mo chasan) "Moccasin, eh? They look useful. Can you make me a pair?" Such a brilliant new idea would be bound to catch on quickly, and they'd be likely to use the foreign word for it as they wouldn't already have one.

But is there any evidence to suggest the Algonquians didn't have shoes before the white man showed up? And if Algonquian shoes were based on European ones, why did Lescarbot (the writer of Nova Francia) consider moccasins so different from European shoes that he used the Algonquian word to describe them? Also, we need independent evidence of pre-1609 Gaelic presence in New England or SE Canada – where is it?

If the reason the word spread so quickly is something other than a previous lack of shoes among Algonquians, why aren't there more Gaelic words in Algonquian?

(d) How likely is this to be just a coincidence? Actually, quite likely. A man named Mark Rosenfelder (webname Zompist) has explained this much better than I could, so I'll simply link you to two articles of his. They're fairly long but his style is nice and colloquial.

On the point of semantic leeway (mentioned in the first article), I should mention that, as far as I can work out using online dictionaries, mo chasan means not "my shoes" but "my feet". This isn't a million miles away, but it's not as close as your opponents were making out.

So basically it's a silly idea. Not prima facie silly, but only when you consider what's known about the history of European/North American contact from other sources. Hope you found all that at least vaguely interesting. I highly recommend the two links above, and Zompist's other linguistics articles – he's a pretty clever chap.

Homework assignment: assess the likelihood that the English exclamation "Smashing!" derives from the Gaelic 's math sin ("this is good").
Camilla likes this


Camilla,  04.06.10 11:00

Nice! I like these linguistics articles.
And I am going to make it my hobby to make the claim that "Smashing" derives from "´s math sin" whenever I am confronted with this type of claim. Although that might make the problem worse.

When I was in Peru, everyone kept trying to convince me that Quechua and English were related. I think their main foundation for that is that several of the numbers from one to twenty are similar.

Tim,  05.06.10 11:59

1 hoq
2 iskay
3 kinsa
4 tawa
5 phisqa
6 soqta
7 qanchis
8 pusaq
9 isq'on
10 chunka

I'm afraid I'm not seeing it. "Phisqa" and "soqta" could conceivably be IE words for "five" and "six", but the rest of them are way out.

11-20 are "ten plus one", "ten plus two", etc... "two-ten".


Low numbers are generally thought to be a good thing to compare between languages as they're very unlikely to be borrowed. But

I think it's worth adding at this stage that surface similarities are not good indicators of genetic relationships between languages. What linguists look for is regular correspondences. This is called the comparative method. A simple example of this would be if you found fifteen words in language X which begin with /tS/ (the sound which English spells as "ch" and Norwegian as "tsj"), fifteen words with similar meanings in language Y that begin with /h/, and corresponding words in language Z that begin with /s/. You can then postulate a proto-language Proto-XYZ which had (for example) /*k/, which changed (at least word-initially) to /tS/ in X, /h/ in Y and /s/ in Z.

/k/ > /tS/, /k/ > /h/ and /k/ > /s/ are well-known common sound changes. But the important thing is the correspondences, not how similar the sounds are. Proto-Indo-European /*dw/ shows up in Armenian as /erk/, e.g. erku "two", from PIE /*dwō/ (see here.) On its own this wouldn't suggest a relationship to anybody sane, but (I'm told) there are a lot more words in Armenian which have /erk/ where PIE had /*dw/, so this is enough to establish that Armenian is an Indo-European language. When you put this in the picture, the "amazing" similarities people sometimes find between languages separated by thousands of miles start to look a lot less relevant.

A fun example of this, by the way: the word for "dog" in the Australian language Mbabaram is... wait for it... "dog". This developed regularly through known sound changes from earlier *gudaga (cf. Yidiny "gudaga" and Dyirbal "guda"). You may recognise Dyirbal as the language that has a noun class for "women, fire and dangerous things".

Camilla,  05.06.10 12:46

Yes. I didn't say they were right. And it may have been numbers other than the first ten. I don't remember. But random people in different parts of the country gave me the same story. I don't know where it originated.

Anders K.,  05.06.10 13:19

I think claims like these don't necessarily need to be based on anything substantial at all. Every now and then, someone will claim that the Romsdal dialect first person pronoun "i" (ee) is obviously derived from English or vice versa, "because it's the same word". The only explanation for this would be that these people don't know how to pronounce a single word in English.

Camilla,  05.06.10 13:27

Indeed. I do love the "i" = "I" argument.
There is something about popular etymologies and people's love of random patterns which is ... endearing? When it doesn't grow annoying.

Tim,  08.06.10 00:58

@Camilla: Of course I didn't think you believed them. I was simply commenting on the complete and utter lack of any similarity between Quechua and English numbers. I am therefore puzzled as to how this claim could have come about.

@Anders: Romsdal i and English I are, of course, derived from the same word in the ancestor language Proto-Germanic, but your point stands: people place far too much importance on spelling. Incidentally, has Camilla told you her story about the girl who, when she and her classmates were asked how they pronounced the first person singular pronoun, announced "I sier jeg"?

Anders K.,  10.06.10 22:40

Ah, children. They're not very good liars.

The Proto-Germanic has spawned as various words as ich, ik, eg, jeg, ja, je, e and æ. In my opinion, the one that sounds the most like the English, is the Sunnmøre "ej", found just across the mountain from Romsdal, however they never make this claim. Of course, volumes could be written on all the other kinds of claims they make.

Another favourite proposed etymology is one I found in a book by guru, religion founder and self-appointed deity Prabhat Ranjan Sarkar. The book, called "The Science of Letters", was one of his approximately 250 books on every possible scientific discipline there is. According to the blurb (written by his zealous followers) a single sentence of his infinite wisdom will give other scientists a basis to work on for centuries. Well, to find out once and for all what this old swami was up to, I sampled through some of his writings, and I did get my suspicion confirmed: The only thing he actually knew anything about was vegetarianism and rail roads. However, the only thing coherent enough to remember from his book on linguistics was his very interesting explanation of the origin of the word mango.

According to Sarkar, the word mango is not, as most dictionaries will claim, derived from the Tamil word for ... well, for the fruit that would have had another English name if the Europeans had seen it somewhere else first. On the contrary, the word was unwittingly coined by the wife of an 18th century British colonial officer, aptly named Mr. Ricecurry (I'd love to know where in Britain he was from). During their stay in India, Mrs. Ricecurry developed a liking to this particular fruit, and whenever she heard a mango seller pass by the house (God knows what he would call out to attract customers) she would send her manservant for some, always saying the same thing: "Man! Go!" (Presumably, "... and buy me one of those sweet, kind-of-orange-y nameless fruits" was somehow implied.) The locals, naturally, believed this was the proper English term for the fruit, adapted it, and it quickly spread across the subcontinent and then the world. Easy peasy.

(In case anyone wonders, the Tamil mānkāy means "fruit or seed (kāy) of the mān tree", and became mango through Portuguese. The reason Mister Lingustic Expert Sarkar omits this is probably that he is a Bengali speaker, in which the word is aama. Bengali and Tamil are not related.)