The Dictionary I Read For Fun
John McWhorter | The New York Times | March 2, 2023
Pablo Delcan
Not long ago, a book that is truly a gorgeous thing, in all senses of the word, was published. It is a third and expanded edition of a full-length dictionary of the Native American language Lakota. And I mean full-length. The New Lakota Dictionary, compiled by the Lakota Language Consortium under the supervision of the linguist Jan Ullrich, is of the physical heft most commonly associated with brands like Webster and American Heritage. (There is also an online version.)
Lakota is primarily spoken in North and South Dakota. The closely related Dakota language is spoken there as well. Of the more than 300 Indigenous languages once spoken in the present-day United States — differing as much as English, Japanese, Hungarian, Thai and Indonesian do — the vast majority now are extinct as spoken languages or are spoken fluently only by people nearing the end of their lives.
Without serious efforts toward revitalization, dozens of them will become extinct a generation from now, according to an estimate in “Ethnologue,” which catalogs languages. Many of the groups, often assisted by linguists, seek to keep the languages spoken in some fashion. Of course, an important step is compiling dictionaries and descriptions of how their grammars work.
Compared with many Indigenous American languages, Lakota is doing rather well, with an estimated 2,000 native speakers remaining, according to the Endangered Languages Project, and this marvelous dictionary may help keep the number of speakers from falling. It gathers over 41,000 words and illustrates them with more than 50,000 sentences, usage notes and collocations.
Lakota is not my language of study, nor are other Native American languages. Yet partly because I am this strange thing called a linguist and partly because I am the kind of linguist who wants to know a little of every language on Earth, I have curled up with this book with a glass of wine countless times over the past couple of months just to savor the cornucopia that this dictionary is.
However, I must admit that one section of the dictionary gives me pause. For this edition, Ullrich has contributed a new section on the language’s grammar, and as expertly composed as it is, it’s hard to miss that Lakota, frankly, is hard!
I’ll give you just a quick sample. In a basic sense, the word for “I” is “wa.” Or more properly, it’s a prefix. So “made it” is “káǧe,” and “I made it” is “wa-káǧe.” If this were a language like English, that’s pretty much all you’d need to know about “I.” But the thing is that with many verbs, you have to jam “wa” into the middle of them. So “I found it” is “iyé-wa-ye,” not “wa-iyéye.” There’s more: If the verb is less about doing something than having an experience, then “I” is said differently, as “ma” instead of “wa,” also often jammed in the middle. “Tired” is “watúkȟa,” and “I am tired” is “wa-má-tukȟa.”
Lakota is like this throughout. But in the grand scheme of things, it is about as hard to learn as most of the world’s 7,000 languages. When it comes to grammar, English is on the easy side. You need just a single suffix, “-s,” to run through the present tense conjugation. There are no suffixes for the past, future or conditional that change for person and number the way they do in, for example, Romance languages. “I” stays the same whether you are doing something or experiencing it, and we certainly don’t plug it into the middle of other words.
But as the world’s languages go, English’s relatively streamlined grammatical nature is by no means the norm. Typically, a language makes you face either a boatload of prefixes and suffixes or, if not, then a lot of tones. To oversimplify, what this means is that a language tends to be like either Russian or Chinese. Lakota is more like Russian.
So if a typical language — i.e., one not like English — isn’t passed on from parents to their children but is learned in school and maybe starting only in the teen years or later, the signal has a way of weakening.
Researchers stipulate that the window on our ability to learn languages with native competence starts closing sometime during adolescence, because of biology or just the fact that you get busier and also shyer about making mistakes.
The result tends to be that beyond older speakers who grew up speaking only the tribal language or learned it alongside English from infancy, a new version of the language develops. Predictably, it is less grammatically elaborate. For example, I surmise that someone learning Lakota in school might not master exactly which verbs you slide the pronoun into, putting it instead at the beginning as one would feel comfortable doing when used to English. Also, vocabulary gets smaller. It is natural to sub in, for example, English words, especially for things most often encountered in English such as laptops and the kiddie goop called slime. You might even often sprinkle in transitional words like “anyway.”
This is the way new generations speak many threatened languages worldwide. It’s been documented in the case of Irish Gaelic, for example. In some parts of Ireland, younger people are certainly speaking it, but to a considerable extent, it could be said to be Irish in English, in which the quirkier Irish rules get flattened out in favor of rendering it the way you would in English. And the truth is that there is nothing wrong with this. We can see it as language changing, especially given that languages are always changing in countless ways, many of them because they’re being used alongside other languages. Languages in the same mouth will mix, unsurprisingly.
For example, my linguist friend Ghil’ad Zuckermann has assisted in the revival of an Australian language, Barngarla, which stopped being spoken in 1964. In his book “Revivalistics,” he notes that New Barngarla, inevitably, does not use some of the things that original Barngarla did. For example, Indigenous Barngarla pronouns were awesomely baroque: “Ngadlaga” means “we two” (but not “we three” or “we four”) and only when used by a mother and child or a man and his sister’s child (not his brother’s) and then only in a sentence that has an object. If there’s no object — as in “We two are sleeping” — then you have to use a different pronoun.
This distinction corresponded, in part, to the nature of kinship in Indigenous Barngarla society. But much has changed in Barngarla lives since then to force them into integration with white Australian society to a large extent, and making distinctions between pronouns like that naturally feels less urgent today. But what they are working with is still Barngarla; it sure isn’t English.
This might seem like a dilution or disintegration, but keep in mind that the language I am writing in is a lexically infected and grammatically streamlined version of Old English. King Alfred would find modern English alternately incomprehensible and barbaric. Many researchers think it got this way mainly because of what Viking invaders did to the language starting in the late eighth century, C.E. They spoke Old Norse, which was related to Old English but different. When they started using Old English, they probably spoke it as well as an American speaks Spanish after a few years of classes — functional, but just.
And they stripped Old English of its harder things, like vast tables of verb conjugations and noun declensions and the meaningless gendering of nouns of the kind that German imposes on “silverware.” It’s why English is the only standard language in Europe (other than, for instance, Basque, Estonian, Finnish, Hungarian and Saami, which are not in the family that all the other languages belong to) that does not assign things to genders.
Then because of the Norman Conquest that followed the Vikings’ invasion, you can scarcely produce an English sentence without recourse to originally French words. Just in that last sentence there are eight (“because,” “Norman,” “conquest,” “invasion,” “scarcely,” “sentence,”“recourse” and “originally”). And the Vikings planted plenty of their own words, too. Without these new users, our word for “take” would be“nim,” our word for “knife” would be “sax,” and we would speak casually of being blithe rather than happy. I doubt anyone sees this as a problem. We speak our English; only the linguist calls it modern English, in salute to there having once been some earlier stage, which to us was a noble but bygone, “Beowulf”-y thing.
Revived languages today are going through a similar process, and the result will be more new versions of languages. A great many languages of the future will be structurally streamlined versions of their original form, but in the end, most of languages’ grammatical doodads are accidental accretions. They creep into a language and pile up over time, and somehow toddlers can wangle them and therefore do. But just as “silverware” doesn’t need to be gendered, full human expression hardly needs eight tones (in Hmong), four gradations of past tense, as in today, yesterday, a little while ago or just now (in Kikongo) or so many prefixes and suffixes that a single verb can appear in 1,502,839 forms (in Archi, a language of the Caucasus Mountains).
Lakota will likely change in the same way that many languages have. And that’s normal. In any case, in this succulent pot roast of a dictionary, the language lives, improvises and even beckons between the book’s covers. I salute Ullrich, the Lakota Language Consortium and the many, many native speakers they collaborated with for 1,420 pages of glory.
