I Can Eat Glass without Hurting Myself

* It also went missing from its original home. Fortunately it has found a new home at Columbia, masquerading as one example of Fun Things To Do with Unicode.

Once you have dealt with hovercraft, can glass be far behind? Here we have not merely a conversation-starter but a genuinely useful piece of information, unaccountably missing from most published phrase books.* Before you travel anywhere, memorize this statement in the appropriate language:

I can eat glass without hurting myself.

It is guaranteed to make locals stand back in awe and treat you with the utmost respect.

plate with glass
Set the Table

So let’s dig up some words. As you can imagine, glass did not play a huge role in traditional Inuit culture. If you’re living a largely nomadic existence in a part of the world where the temperature drops below zero—Fahrenheit—in November, and stays there until further notice, some things have to be classified as non-essentials. But it’s an accommo­dating language; look only at the multitude of words for “frog”. The dictionary comes through with ᐊᓕᒍᖅ, aliguq. The word started out meaning “quartz”, which is not as picky about temperature; “glass” is an obvious extension. More recently, the word has also come to mean “diamond”. But that’s taking us into Let’s Not Go There territory.

* Raw caribou: Sure, bring it on. Walrus heart: . . . Uhm, is this one of those things I gotta eat for the Vitamin C? Polar bear liver: Used to be an excellent way to get rid of unwanted mission­aries, but most people are wise to it now. Live warble-fly larvae**: No, but thank you for offering.

** I don’t make this stuff up, y’know, although possibly the original author—or his teenaged informant—did.

Nobody said he was the world’s leading authority on punctuation.

There are probably lots of words for “eat”. I’m going to use ᓂᕆ (niri-) on the rock-solid grounds that I already know it. As a bonus, this is an intransitive verb root. If you’re dining on some­thing particularly noteworthy, the verb can become transitive or specific: you’re eating some­thing. If you prefer not to inquire too closely,* it’s intransitive or non-specific, but nothing else has changed. No danger that it will come out meaning “I chewed off my arm” when all you meant to say was that you leaned your elbow on the table.

Overdue digression: What’s with that hyphen at the end of every verb? Well, let’s quote the world’s leading authority again:

Verb Roots cannot exist on their own; they always need an ending. So I always write Verb Roots with a dash after them . . . .

If you look up a verb in the dictionary, you will not find a bare niri. You cannot say “to eat”, because Inuktitut verbs don’t have infini­tives. Our friend the Reverend Peck thought they did. In fact, most missionary linguists—including the Moravians, who generally had better sense—insisted on using the word. But the forms they call infinitives turn out on closer inspection to be “participles”. Blame it on Kleinschmidt. He’s got a convoluted argument about why the word “parti­ciple” is wrong—while failing utterly to notice that “infinitive” is even wronger. But at least calling it a participle stops you from expecting an uninflected form. Nothing in Inuktitut is unin­flected. Well, maybe a few interjections. Even Inuktitut does not expect you to put a-aaq! (“Ouch!”) into the appro­priate gramma­tical form based on the person and number of the thing you just stubbed your toe on.

Lacking a handy infinitive, the dictionary will instead give you

ᓂᕆᔪᖅ nirijuq (vb. intrans.) he eats


ᓂᕆᔭᖓ nirijanga (vb. trans.) he eats it

Or possibly nirivuq and nirivaa, depending on the dialectal preferences of the dictionarist. And if you are lucky enough to lay hands on Alex Spalding’s dictionary, the “he” and “him” forms will be varied by “she” and “her” according to an arcane, mysterious and sometimes quite glorious formula known only to the author. As, for example,

ᓇᒡᓕᒍᓱᒃᑐᖅ nagligusuktuq & ᓇᒡᓕᒋᕚ nagligivaa (trans.) he loves her

followed by derived forms such as

. . . they love each other

. . . he loves himself

. . . she shows sudden affection or feeling for him (grabs or kisses him suddenly or unexpectedly)

And I can’t leave Spalding’s dictionary without quoting one of my favorite entries:

ᐄᖅ! iiq! (excl. of distaste or annoyance after finding one has gotten wet paint on one’s trousers or that the dog has puked in the living room) yukk!

Now that dear Dr. Johnson is no longer with us, we are not often treated to autobiographical dictionary entries.

Well, that was fun. Back to business.

Hurting myself was unexpectedly troublesome. Possibly it’s another of those luxuries some people can’t afford. I had to go with ᓱᕋᖅ (suraq-) or—your choice—ᓱᕋᒃ (surak-). This may be linguistic overkill, since the dictionary tells me it means “destroy irreparably”. Sure, the glass didn’t do me any good, but after a few months in Intensive Care I’ll be managing my esophageal tube like a pro. Unlike niri-, but like the tatat- that played such a big part in your hover­craft’s recent existence, this is a transitive verb. And that’s just what we want. I’m not planning to hurt anyone else, but I am planning on getting hurt myself. Or rather, on not getting hurt. But that’s a grammatical trifle.

plate with glass
Prepare the Meal

1. Glass

* In quotes because technically “subject” and “object” are grammatical terms, while “agent” and “patient” are real-life terms. The agent is the person who is actually, physically eating the glass, regardless of whether the sentence describing the event is active (“I eat glass”), passive (“Glass is eaten by me”), or some other construction entirely (“The eating of glass is char­acter­istic of me”).

We’ll start with the glass. You can never go wrong putting the patient (“object”)* of a verb first. Thanks to the verb involved, we can go either way: glass-mik with intransitive ending, or glass alone with transitive ending. Let’s make it transitive, though the glass may not be specific enough to warrant it. In fact, the whole point of the utterance is that you can eat any glass, anywhere.

aliguq (patient of specific verb)


aliguq + mik = aligurmik (patient of non-specific verb)

2. Eat It

Transitive or not, this is obviously not a simple matter of nirijara (“I eat glass”). We could go with a conditional (“if I eat glass...”) or a participle (“while eating glass...”), or a frequen­tative (“when­ever I eat glass...”). Let’s stress the frequency of the activity.

Further digression: I was recently hit with a whole flurry of hitherto unsuspected verb modes. There was the Dubitative niringmangaakku (“whether I eat glass...”), the Negative Participle nirinagu (“while not eating glass...”), and most dramatically the Frequen­tative nirijaraangakku (“whenever I eat glass...”). Those are all conjugations in their own right, with nary an affix in sight.

* Leading to a further choice between Option A—an affix in r deleting the preceding q—or Option B—an affix in g fusing with q to become r. To identify the underlying form you have to be either much better at internal recon­struction than I am, or spend more time studying related languages than most people’s life expectancy will allow.

Unless, that is, you go with Interpretation B. In this version, there’s a Frequentative affix (-jaraanga-, -taraanga-, -kaanga- and -raanga-*) and a Dubitative affix (-mangaaq- with assorted forms of inter­polation and assimilation resulting in initial -ngm-, -mm-, -ngm- and -rm-). Each is then followed by—hang on while I look this up—“personal pronoun suffixes in the Genitive”. This information might be more useful to me if I had some remote glimmering of an idea what the said “personal pronoun suffixes” are. But that’s another problem for another day.

Those Personal Pronoun Suffixes are for the intransitive or non-specific forms. The book is silent about what happens with the transitive or specific. But not, unfortunately, because the verbs them­selves are silent. Sort them out and you’ll find a further 58 endings, just like the Becausa­tive and Condi­tional. In fact, they generally rhyme.

* Assuming for the sake of discussion that you know the difference between “lay” and “lie”.

You were wondering when I’d get back to this, weren’t you? The Becausative mood is the one that means . . . drumroll . . . “Because.” Old-school linguists tended to label it Subjunctive. Or Subjunc­tive Form A, while the Conditional (“If”) was Subjunc­tive Form B. For a good six months I went around merrily calling it the Causa­tive, before remembering that that’s something entirely different. The real Causative—which you can’t rename, because the term is solidly established—means “cause someone to do something”. That’s where you get words like “lay”,* “fell” and “set”, which mean “cause to lie”, “cause to fall” and “cause to sit”, respec­tively.

In some languages the Causative is a productive form. That is, you can take any verb, apply the appropriate rule, and it becomes causative. In English we’re limited to pre-existing words created long ago by Old English or possibly Proto-Germanic ablauts. And in Inuktitut the job is predictably done by an affix, most often listed as ᑎᑦ. Transliteration omitted, or the search engines will go haywire.

The word “Becausative” is, admittedly, not exactly English. Or rather, not exactly Latin. Put it in the same bin as television, sociology and all those other words that make linguistic purists foam at the mouth. It starts with the strictly English prefix be-, changes gears with cause—Latin by birth, but fully naturalized—and winds up with the strictly Latin suffix -tive (lightly modified from its original -tivus). But it conveys the necessary information and, as a bonus, it’s cute. I swiped it from Inuktitut Computing, which in turn swiped it from Mallon. You could do worse.

That “58 endings” sounds like a lot, especially when you add the intransitive forms to bring the total up to 70. But consider that each verb could have 120 endings in each mood: 12 intransitive plus 12 × 9 transitive. Or 90 in total, for the three moods that don’t have any truck with a fourth person. From that perspective, 70 or 52 doesn’t sound so bad.

* Returning to the technical theme: a parti­ciple is an adjec­tive made from a verb—and Inuktitut does not, technically, have gramma­tical adjectives. They’re verbs, thank you very much, and prefer to stay that way. Sometimes participles are called Conjunc­tives instead.

The three kinds of “participle”* are best of all. Not only are their endings identical after the first syllable, with a supple­mentary syllable doing the job of the fourth person. They also do a neat toggle whereby each intransitive form for a given agent (“subject”) becomes the transitive forms—all of them—for everything with that patient (“object”). While I’m looking around in a general sort of way, everyone turns around and looks at me. Same word. Grand total for the three variants: well under twenty endings plus a couple of simple rules, instead of the expected 360 or at least 200-odd.

nirijaraangakku (specific)


nirijaraangama (non-specific)

3. Get Hurt

If I intended to hurt myself, a simple ᓱᕋᖅᑐᖓ (suraqtunga) would do. Attaching intransitive endings to a transitive verb can be dangerous. So let’s slip in a quick -nngit- to wipe out the harm.

In English, the whole thing starts with “I can”. But it doesn’t really apply to the immedi­ately following word “eat”—anyone can eat glass if they don’t care what happens next—it refers to the whole sentence. It’s the ability not to get hurt that is noteworthy. So let’s grab an “ability” affix and shove it in. Happily I was able to lay my hands on one that starts with t, -tunnaq-, minimizing the damage done to the preceding -nngit-. It even creates the illusion of restoring the q from the root verb that got eaten by that same -nngit-. In Inuktitut you can’t lurk in wait for the next affix to come along; you’re only allowed to maul the one that was directly ahead of you in line.

ᓱᕋᖅ + ᙱᑦ + ᑐᓐᓇᖅ + ᑐᖓ = ᓱᕋᙱᑦᑐᓐᓇᖅᑐᖓ
suraq + nngit + tunnaq + tunga = suranngittunnaqtunga

plate with glass
Chow Down

ᐊᓕᒍᖅ ᓂᕆᔭᕌᖓᒃᑯ ᓱᕋᙱᑦᑐᓐᓇᖅᑐᖓ
aliguq nirijaraangakku suranngittunnaqtunga.

or, if you prefer not to get too specific:

ᐊᓕᒍᕐᒥᒃ ᓂᕆᔭᕌᖓᒪ ᓱᕋᙱᑦᑐᓐᓇᖅᑐᖓ
aligurmik nirijaraangama suranngittunnaqtunga.

Mm-mm, yum! Didn’t feel a thing.

mouse looking at an apple with a worm in it
I Can Eat Live Warble-Fly Larvae
with a Straight Face

Readers in Baker Lake and points east may breathe easy. There is not one word of Inuktitut in this section.

* For readers who are concerned about this kind of thing: In spite of its recent date, this book is out of copyright in the United States because it was originally published without a copy­right notice. Universities tend to forget this detail fairly often—except when they go to the other extreme and slap copyright notices on material that existed before the University founder was even a gleam in his mother’s eye. So Americans can get the whole thing at Project Guten­berg. People in the rest of the world should note that the author died in 1972, and act accordingly.

We now detour to The Barren Ground Caribou of Keewatin by Francis Harper, 1955.* You will deduce from the title that the author was neither a linguist nor an anthro­pologist but a wild­life biologist—and apparently a pretty good one. The research for this book was done when he was in his 60’s. So his career spanned the development of wildlife biology from its original form—“It’s the last breeding pair on the planet? Shoot ’em quick, so we can stuff and mount them!”—to the beginnings of its current “observe them in their natural state” form. The author himself admits that “We know as yet extremely little concerning the movements of individual Caribou”, with footnote:

One means of gathering information on this subject would be to capture fawns as they swim across lakes or wide rivers on the autumn migration, then to affix numbered metal tags to their ears and to release them in time to rejoin their mothers. This would simply be a modification of the leg-banding method that has proved so highly successful in the study of bird migration. It would also be particularly useful in studies on age and growth.

* I lied about no Inuktitut.

Promising line of thought, eh? Overall the author seems to have been a pretty decent guy—except when he became a godfor­saken idiot, as white males tended to do in 1955. But there’s no need to go there. What intrigues me is his account of the warble-fly larvae. No relation to the “nostril flies” (his term) which we will shortly come to know and love as the taggiuk. Warble fly larvae are generalized ectoparasites. In fact, they are so ubiquitous that an almost-grown male caribout is called a ᖁᒪᓕᒃ (qumalik): something that has qumait.*

According to our author,

The larvae of the warble fly (Oedemagena tarandi), found beneath the skin of the Caribou, are relished by the Eskimos, being eaten apparently while alive and raw. The Eskimo boy of our camp continued this practice after his little sister had given it up.

You don’t have to take his unsupported word for it, either. The annotated bibliography includes four different sources, ranging from 1795 to 1927, that say the same thing. And maybe they’re right.

On the one hand: Five published sources, at least one of them based on personal observation. There is no inherent reason why someone should not eat warble-fly larvae. I happen to come from a non-bug-eating culture. I have no information one way or the other about this aspect of the Inuit culture. But we can definitely say that theirs was not an environment where dietary fastidi­ousness was a survival trait.

On the other hand: Our author’s direct observational source was a fifteen-year-old boy. (We know this from elsewhere in the text. There are even photographs.) It is all too easy to picture someone from this particular demographic group testing to see just how credulous the American visitor could be.

There you have it. Two different, mutually exclusive, fully plausible explanations of the same observed phenomenon.


At this point, I am sorry to say that my willpower gave out and I went to look it up. This presently brought me to the 2003 Presidential Address of the American Society of Para­sito­logists, “Flavor Buds and Other Delights”, delivered in Halifax, Nova Scotia, by Robin M. Overstreet of the University of Southern Mississippi. Since it was originally presented orally, there is no Abstract to plow through. Instead we get a closing Summary. In full:

Using good sense and some biological information, one can enjoy a delightful morsel or enhanced meal from a variety of parasites, either raw or cooked.

Bon Appétit!

* In academia, “some guy told me” is not hearsay. It is a Personal Communi­cation.

The author did not personally sample warble-fly larvae, but he talked to someone who did.* In parasitologist-speak, “they tasted quite good, with the hemolymph providing a saline sensation and the remainder of the warble having a nutlike flavor.” We also learn that the larvae can be cooked, but they aren’t as tasty that way.

Before moving on to the next creepy-crawly, our author tells us that the Iñupiat word for “warble fly”, kumak, also means “louse” (the kind that lives in your own hair). This naturally sent me racing to the dictionary, where I found, on the one hand,

ᖁᒪᒃ qumak (n. zoolog.) caribou back worm or parasite

and on the other,

ᑯᒪᒃ kumak (n. zoolog.) louse; or (gen.) bug

* Ulirnaisigutiit: an Inuktitut-English dictionary of Northern Quebec, Labrador and Eastern Arctic Dialects, by Lucien Schneider.

In the US, the price of this book ranges from $80 to $240 used. On the other side of the border, you might manage a new copy for around C$90. If you find it at $80, do not sleep on the decision . . . unless you are planning a remake of Sleeping Beauty in real time.

When I first read this, I naturally assumed that our source’s source was but one more in the long line of people who couldn’t hear the differ­ence between k and q. I’ve since learned that the two words really are the same in Iñupiat.

They are also—according to yet another dictionary*—the same at the other end of the dialect continuum, in Labrador and Quebec. But once you get to Labrador, all phonetic bets are off. The dialect has both k and q, but their arrange­ment seems to have no relationship to anyone else’s. Can’t blame it on orthography, either. This was the domain of the Mora­vians, who could hear the difference—and invented the ĸ (“kra”) to prove it. So if someone tells you they came face to face with a nanuk, don’t jump to the conclusion that they were too frightened to hear the q. It may have been a Labra­dorian bear.

There is no shortage of lice-with-a-k in Greenland. But Kalaallisut won’t admit to the existence of warble flies, nor yet their larvae, so don’t look for any bet-settling from that quarter.

Did two similar words with similar meanings collapse into one—or did one word with two slightly different meanings diverge into two? If the linguists can drag themselves away from the saline sensation and nutlike flavor, they might like to chew on the question.

* Something along the lines of “your aid to under­standing”, in case anyone wondered. The verb root—in Inuktitut, there is always a verb—is ulirnaiq, “under­stand or comprehend”.

Wait, there’s more to chew on. To make up for conflating ᖁᒪᐃᑦ (qumait, warble-fly larvae) with ᑯᒪᐃᑦ (kumait, lice), the Ulirnaisigutiit* also offers up

ᖁᒪᖅ qumaq (pl. -it) intestinal worm, white tape worm

or, for variety’s sake,

worm of pools and lakes

These q-laden wigglies also exist in Greenland. Elsewhere, people avoid the issue by calling them ᖁᐊᖅᑐᖅ (quaqtuq).

Watch this space. I have only to find a ᑯᒪᖅ (kumaq) permutation and we will have the complete set.


Darn it all. I was really rooting for that fifteen-year-old’s powers of deception.