=== Odd Spellings ===
“these People are Singelir from the Rest of thare Nighbors. thay
Speake a Hard un Corth Langwige Scarst to be Larnt By Eney People. thay will
not aSosheat with or Convars with the other tribes Nor Intermarey among them”
— Peter Pond,
fur trader, ca 1800
A long time ago I watched a
documentary about the Powhatan and the settlement of Jamestown. The producers of the documentary had felt the
need to use more authentic Native American names for the Powhatan leaders, so
instead of “Chief Powhatan” the video scrupulously called him by his Powhatan
name: Wahunsenacawh. This is one of two
common spellings of his name, the other being Wahunsonacock—it makes no
difference which you prefer. The
Jamestown settlers themselves were not very precise in how they wrote down
these Indian names. But the producers of
this documentary wanted to be authentic, and so they instructed the
narrator to say it with, what they felt to be, the correct and respectful
pronunciation of “wah-hoon-seh-NAH-kuh-wuh” /wa.ˈhun.sɛ.ˈna.kə.wə/.
This is in all likelihood not
the correct way to say this name. But
more important than the fact that they said it wrong, is the why of
it. There are two ways to look at old
names like these. Way #1 is to treat
them like foreign names—like Toshiro Mifune or Omar Sharif—and say them with an
“international” pronunciation, enunciating the vowels according to their
Spanish or Italian values: <u> as “oo”, <e> as “ay” et cetera. Way #2 is to read them as though they’re
English names and give the vowels their English values, the kind we were taught
in school: short ŭ as in “cut”, long ū as in “cute”, long ā as it “wait”,
etc. The latter is used in certain
grandfathered-in Latin phrases such “semper fi” and “habeas corpus”, which believe
it or not is how Latin used to be pronounced by English classicists. The reason for the discrepancy is the English
Great Vowel Shift—during the Middle Ages, the English language still wrote and
pronounced its vowels as Spanish and Italian do today.
The problem is that the Jamestown
settlers were not Geoffrey Chaucer, they were not Italians, and they were not
modern linguists. Let me illustrate with
another example. The Massawomeck are a
“lost tribe” of North America who once inhabited the headwaters of the Ohio
River in western Pennsylvania. They
feature briefly in the records of the Jamestown colony and the Dutch colony of
New Netherlands, but were soon after destroyed, absorbed, or run off by the
Susquehannock. They are a pretty obscure
group and so their name is no longer in common use, and you almost never hear
it spoken out loud. I have heard it “in
the wild” in fact only once, in the Disney film Pocahontas of all
things. In the opening scene, when the
aforementioned Wahunsonacock (not so called in the film) arrives at the village
of Werowocomoco (ditto), he announces to the resident Powhatan that his war
party has defeated the Massawomecks. The
voice actor (Russell Means) pronounces it more or less how you’d expect:
“mass-uh-WOE-meck ~ /mæsəˈwomɛk/.
Why is this wrong? Because linguistically we can identify the
name Massawomeck as meaning “people of the big valley” in Virginia
Algonquian—phonetically it was something like məhsēwamak. To understand why the Jamestonians spelled
this as “Massawomeck” you need only imagine that word as if written with little
Webster’s-style pronunciation guide vowel diacritics: \mässāwŏməck\ or
“mah-say-wah-muck” as the Powhatan name was adapted to 17th century
English. In other words the vowel in the
second syllable was “long ay”, the third syllable “short ŏ”, and so on. Over time, though, this pronunciation was
lost as the name fell out of use, and has been replaced with a new
pronunciation matching what it looks like. This is called a spelling pronunciation.
This is not to say that the movie got
it wrong. None of the spelling
pronunciations (and other mutations) in this post are wrong exactly,
since that is just how they are said now.
It is what it is; languages change; old forms are lost; new forms replace
them—life goes on. Russell Means was
right: that is how “Massawomeck” should probably be pronounced in modern
English, which after all is not Jacobian English. I assume it’s how most historians of colonial
Virginia say it. Insisting on saying it
in the old way is a bit like insisting that the characters in that film all
pronounce the “k” in the word “knife”.
Sure, that’s how it was pronounced at the time... but who carse?
The difference, I think, is that the
producers of the Jamestown documentary clearly were trying to give it an
original, authentic pronunciation, and in this endeavor they failed. They didn’t quite know what they were
doing. “Wahunsenacawh” looks like it
should be /wahunsɛnakəwa/ only if we treat it as a foreign name written in a
technical orthography, but doing it that way is backwards. It’s the mistake of thinking that written
language comes before the spoken.
Powhatan was not generally a written language (and on occasions when it
was written down, it wasn’t in a modern phonetic orthography). First came the Powhatan spoken form, which
was then imitated by the settlers as an English spoken form, which was then
given a written spelling according to contemporary orthographic standards. And it is those standards—augmented by comparative
Algonquian linguistics—which you must observe if you really wish to be
authentic [note A].
But I get it, I know why the doc said
it the way it did. Something about this
colonial way of spelling just feels... wrong.
“Massawomeck” doesn’t look like it should have a “long ay” in the second
syllable or a “short ŏ” in the third.
It’s hard to put into words why, it just goes against our
orthographic sensibilities. It doesn’t
just feel wrong, it feels stupid, like Homer Simpson saying “gym” like /gaim/. Or like a kindergartender who hasn’t yet
learned that final “e” is silent in words like “Dave”.
After all how do you pronounce
“Dave”? That’s a fun example, on account
of the most famous Dave in American frontier history: Davy Crockett. A lot of people say that Mr Crockett himself
(who was a congressman thank you very much!) had too much self-respect
to ever call himself Davy—that he insisted on the proper name of David. Indeed, that is what he calls himself in his
autobiography, A Narrative of the Life of David Crockett (my edition
carries the less dignified retitle King of the Wild Frontier: An
Autobiography by Davy Crockett). But
the circumstantial evidence is that he did sometimes call himself Davy...
except you might easily miss it, because he spelled it “Dave”. As in \dāvē\, as in “day-vee”, as in you do
pronounce the final long “e”. This at
least is suggested by Paul Andrew Hutton in The Undiscovered Country
(the history book, not the Star Trek film).
A better example is the name of
Iowa. This was originally meant to be
pronounced like Ioway—a spelling which is still used when referring to the
eponymous Indian tribe from Iowa (personally I like that there’s a way to
distinguish the two). In other words it
was said with all three vowels having their Webster’s-style long values: \Īōwā\
- “EYE-oh-way”. This is how my
grandfather used to say it. But for me
and my generation (and I assume you as well), it just feels wrong to put a
single “long ā” at the end of a word. We
can do it for “long ē” in words like psyche; we can do it for “long ī” in words
like alibi. But we don’t do it with
word-final “ā”. At least not
anymore—watch the opening scene of 1942’s Casablanca though, and hear
the narrator pronounce visa as “vee-zay”.
Sometimes this even goes the opposite
direction. The name “Kiowa” has always
had more or less the pronunciation it has today, as indicated by early Spanish
spellings of the name and its form in various other Native American languages. But in the old days, occasionally people
pronounced it like “Kioway”—this presumably was because of a mistaken
assumption that it ought to rhyme with “Iowa(y)”. Robert Duvall says it like “Kioway” in Lonesome
Dove, for instance, though whether as a conscious character affectation or
because Duvall himself (who was kind of an old-fashioned guy) pronounced it
like that, I don’t know.
Minding the spoken-written distinction
like this can help you avoid certain mistakes.
I’ve written before how the “Quiohuan” tribe spoken of in the Spanish
and French records of the southwest were not the Kiowa because—among
other reasons—that theory relies upon assuming an English pronunciation of the
letter “i” in those names. A similar
fallacy is the idea that the “Chien” Indians recorded from the Minnesota area
were the Cheyenne on account of “chien” being French for “dog” and the Cheyenne
having a warrior fraternity called the Dog Soldiers. The Chien were the Cheyenne, yes, but only
because “Chien” is an anomalous English spelling of the name (with a “long ī”),
and it has nothing to do with dogs.
The Last of the Mohicans features a character named Chingachkook whose name—in the film
version at least—is pronounced /čɪŋgəčgʊk/.
James Fenimore Cooper found the word in an old Lenape dictionary, it
means “Big Snake”. But that dictionary
had been written by Moravian missionaries, and it used a German-based
orthography which employed the “ch” digraph to write the velar fricative /x/
(the sound of Scottish loch, Spanish baja, German achtung). A modern phonetic orthography would render
this word as “xinkɔxkook”. If this name
had ever been borrowed directly into English then it would probably have
sounded like “Kingakuck” or something like that. Whether James Fenimore Cooper himself was
aware of this I don’t know, but the actor Russell Means (what, him again??)
evidently was not.
The same kind of mistake happens in
the more recent French and Indian War film Young Washington, which
features a character (and real life person) named Half-King Tanacharison. In the film this is pronounced /tænəčəˈrisən/
~ “tă-nuh-tcha-REE-sin” but you can tell that this is wrong by comparing the
various ways it was written by contemporary English (Tanaghrisson,
Tanacharisson, etc) and French (Tanahisson, Thaninhison, etc) [note B]. I haven’t found a linguistic analysis of this
particular name, but you can infer from these spellings that in the original
Iroquoian it had /hr/ in the middle.
Clusters like/hr/ feel awkward to English speakers—we don’t like having
/h/ after a vowel like this—so in old names such as this one the /h/ was
occasionally written with the guttural-looking “gh” for emphasis. You see this, for instance, in Joseph Brant’s
translations of the Bible into Mohawk.
The word for “Bible” itself Brant wrote as “kaghyadoghseradogeaghty”,
corresponding to modern Mohawk kahiatonhseratokénhti.
 |
| Genesis 1 in Joseph Brant's 1787 translation (left) and in Harvey Satewas Gabriel's 2023 translation (right). |
The records of 17th century Seneca
show it to have been a “normal” Iroquoian language, however over the following
two centuries it experienced some pretty drastic sound changes. One of these changes was that the cluster
/hr/ first was reduced to /r/ and then disappeared entirely. Tanacharison is suspected to have been a
Seneca (possibly by adoption) so it’s not unlikely his name was Seneca, and
thus you can observe that some spellings preserve an older form of it (e.g. Tanacharison)
while others have the newer form (Tanahisson).
The former may have come from a conservative Seneca dialect which hadn’t
yet undergone these sound changes, or maybe it was the same name rendered in a
related Iroquois language. Meanwhile
amongst the English, that /h/ situated awkwardly after a vowel was hardened
into a “k” resulting in a pronunciation probably more like /tanak(ə)risən/. This is the more historically accurate
pronunciation of the name—at least the English version of it.
There are other examples of /h/
hardening to a /k/ after a vowel.
“Schenectady” comes from the Mohawk word skahnehtá:ti meaning
“beyond the pine trees” and uses both English strategies for dealing with pesky
aitches: the first /hn/ cluster was turned into /n/, and the second /ht/
cluster was hardened into /kt/. Other
examples show us that at least some English speakers, though, were trying to
keep those h’s intact. “Caughnawaga” comes
from the Mohawk kahnawà:ke meaning “place of the rapids”; it uses the
same <gh> spelling used by Joseph Brant.
You can also tell from the Mohawk that it was pronounced like
“Caughnawagay”, so it’s another example of the “Ioway” phenomenon. Both details have been lost in the modern
English pronunciation “kah-nah-WAH-ga”.
Moving further west... There is a
place in Minnesota called Mendota, whose name in the Dakota language is Bdóte. There’s a noticeable similarity between these
two names, but the truth is that “Mendota” is more than just an imperfect
imitation of the original Bdóte—in fact the original version of
the name didn’t much resemble either the modern English or Dakota
versions. Written sources from the early
1800s attest that the sound which is /d/ in modern Dakota was originally a
prenasalized /ⁿd/. This was true even in
the name of the tribe itself, which Joseph Nicollet recorded was in his time
pronounced like “Ndakotah”. Another
example is from the journals of Lewis & Clark, where the Mdewakanton band
of Sioux are referred to as the “Mindawarcarton”—meaning that at the time, it
was pronounced in Dakota as Mⁿdewakanton (or technically: Mⁿdewákhaŋthuŋwaŋ).
In modern Dakota, native speakers of
the Mdewakanton dialect call themselves Bdewákhaŋthuŋwaŋ, but capable
linguists in the 19th century recorded this form as Mdewákhaŋthuŋwaŋ. This is due to yet another sound change which
Dakota underwent during the early 1900s: the consonant cluster which was
originally /mⁿd/ and later /md/ was evolved yet again into /bd/.
So we can see now what happened with
Mendota and Bdóte. In the early 1800s
this name was pronounced by both the Anglo-Americans and the Sioux as /m(ə)ndote/,
which the former wrote as “Mendota” (as in dictionary-style: \mĕndōtā\). As happened with Iowa and Caughnawaga, the
“ay” vowel was later lost so that the name is now pronounced “men-DOH-ta”. Meanwhile in the Dakota language, the old Mⁿdóte
was altered via regular sound change to Mdóte by the mid-19th century
and then again to Bdóte by the mid-20th.
So the two languages carried the name off in different directions, even
though originally they were quite similar.
Another example, this one varying
between dialects rather than over time.
The Cree of historic times had various names for their enemies to the
west, principally the Blackfeet whom the speakers of the Plains Cree dialect
referred to as Ayahčiyiniw.
Because of a pattern you find among the Cree dialects (and of Ojibwe),
words which have an “l” in one dialect will instead have an “n” in another, a “y” in another, a “th” in another, and an "r" in another.
Thus early documents referring to the Blackfoot as the “Archithinue” are
representing this name according to the Woods Cree dialect: Ayahčithiniw
(or Ayahčiðiniw). Another
spelling of this name is “Iactchejlini” which at first doesn’t look like the
same name at all—but in fact it comes from an “l”-dialect (this time of Ojibwe)
variant: Ayahčilini. The spelling
of <Iactchejlini> has the letter “j” standing as an alternate for “i”;
the initial letter “I” standing for its schoolhouse-long value of “eye”; and
the awkward-in-English /hč/ cluster apparently being hardened into /kč/,
similar to what happened with Tanacharisson and Schanectady.
Ayahčiyiniw (et var) is sometimes translated as “slave”. I’m not sure how accurate this is—the Handbook
of North American Indians suggests that it’s not—but the Cree did refer to
their Athabaskan-speaking enemies to the north using other words for
slave. The name of a foreign ethnic
group also being a word for slave has ample precedent in world history, most
notably in the origin of the word “slave” itself which has the same root as the
word “Slav”. But the English in Canada,
who tended to take the side of their Cree partners in the fur trade, referred
to these Athabaskans as the “Slaves” as well.
This is where Great Slave Lake in Canada gets its name.
But an unexpected detail. There is an Athabaskan group in Canada which
is still often called the Slave tribe, though in this case the pronunciation of
their name isn’t what you would expect.
It’s pronounced with the dictionary-style phonetics of \slāvē\ with a
“long ē” in the second syllable.
Sometimes you see it written as “Slavey” to make this pronunciation more
explicit. But why does it have that
pronunciation anyway, if they’re named for being the victims of Cree slaving
raids? Because the “ey” ending is
actually based on the French suffix –ais, as in Francais, Anglais,
Japonais. So, they took the
English translation of a Cree word, slapped a French suffix on it, and then
gave that an English spelling which now has a counterintuitive
pronunciation.
And on that note, this stuff happens
in Spanish and French too. Sometimes you
get a name like “Nez Perce” where the pronunciation has been anglified and spelling-pronunciation-ified
from its original “nay pehr-say”—but you shouldn’t always assume this is
what’s going on, if a name looks like it has a French spelling but isn’t
pronounced as a Parisian would. Take Des
Moines for instance. Have you ever wondered
why this isn’t pronounced “de mwaan”?
It’s not a mistake: it originally means “[place] at the mooyinkweena”
referring to a tribe referred to by the Miami-Illinois Indians in their own
language as (for reasons that aren’t quite clear) the “Shitfaces”. In general, the variety of French spoken in
colonial America didn’t have the modern pronunciation of <oi> as “wah” as
in C’est moi—instead it was like “weh” or “way”. Hence the pronunciation of other American
French terms like travois. The case of
“Iroquois” seems a bit ambiguous to me, however, since it likely derives from a
Basque trade jargon term hilokoa or “killers”.
French also differs from English and
Spanish in that it uses <ch> to write the sound of “sh”. There is a small tribe in Louisiana called
the Chitimacha whose name the locals now pronounce like /čɪtiˈmačə/
“chitty-MAH-cha” or /čɪtəˈmačə/ “chitta-MAH-cha”. How it was pronounced in the 1700s is more
difficult to work out—for one it doesn’t seem to be a native Chitimacha
expression itself—but in the early 1900s it was pronounced closer to how you’d
expect if you take it to have French spelling, /šɪtiməˈšɔ/
“shitty-ma-SHAW”. This at least is how
Mary Haas pronounced it; she was married to Morris Swadesh, the linguist who
documented the language from its final speakers, so she would know. According to scuttlebutt on the old Siouan
Mailing List, the modern pronunciation with an English affricate “ch” is not
just a happenstance spelling pronunciation—it was a deliberate change made by
the now English-speaking Chitimachas, because they didn’t want their tribal
name to begin with “shitty”.
But things aren’t always this
simple. The French could spell a name
with <tch> if they wanted to really emphasize that it had the sound of a
hard affricate like English “ch”. An
early French spelling of <Tchioutimachus> thus implies that Chitimacha
was originally said like “ch”. And if
the name does come—as has been claimed—from Choctaw šotiʼ-imą́šah, then
that means the local pronunciation of this name has gone from having an initial
“sh” to an initial “ch” and back to “sh” and then back again to “ch”. On the other hand, colonial French spelling
isn’t always dependable like this. For
some reason the name of Natchitoches is also written with a hard-looking
<tch>, despite the fact that it comes from the Caddo word nashit’ush
with a soft “sh”. There appear to be
depths to this that I don’t grasp.
A similar thing happened with the
name of the Hopi. Prior to the 1890s
this tribe was generally referred to as the Moqui—and not just by outsiders
either: their own name for themselves was Moqui. The original pronunciation of this was as
Mokwi which, when the Spaniards arrived in the 16th century, they wrote as
“Moqui”. In the conventions of modern
Spanish, this would be written as “Mocui” because the alternative (“Moqui”)
implies a pronunciation of Moki.
But in the old days Spanish used <qu> to write kw, as Latin
and English do—so words like cualquiera were written as qualquiera. What this meant for the Moqui was that their
name as written was ambiguous, and so... eventually... the Spanish adopted the spelling
pronunciation of Moki. This
pronunciation was later taken up by Americans as well. Problem is, in the tribal language the word moki
means “dead”, and some folks took issue with that. So in the early 1900s, the anthropologist
Jesse Walter Fewkes suggested the tribe change their name to Hopi, and the rest
is history [note C].
 |
| From an 18th century edition of Cabeza de Vaca's Naufragios. Note the use of "quatro" instead of modern "cuatro". |
This means also that in the recent
film Horizon (which takes place in 1863), when Sam Worthington’s
character refers to the “Hopi” this is a historical error.
 |
| Article on the Moqui/Hopi tribe, from Drake's 1884 abridgment of Schoolcraft. |
Mainstream info sources will just
straight up lie to you about this, by the way.
Web articles and museum placards will claim that “Moqui” was a
derogatory term applied to the Hopi by their enemies because it means “the dead
ones”. This is false. Moqui was in fact the Hopi’s name for
themselves, it’s just that it matters how you pronounce it. Others will also blame the Americans for
originally misreading “moqui” as /moki/ though this seems unlikely to me, as it
is a Spanish spelling pronunciation.
This is parcell with a tendency to insist that all traditional tribonyms
are derogatory in origin—as examples like “Shitfaces” and "Slave(y)" show this sometimes is
true, but certainly not always, and when it is it usually has origins with a
rival neighboring tribe.
Words like qualquiera are just
the tip of the iceberg though. The
Spanish language has gone through a few spelling and pronunciation changes over
the past five centuries which make reading colonial names a bit trickier than
their French and English counterparts.
The simplified version is this:
Back in the 15th century or so, the letter x stood for the sound “sh”
and the letter j (or g before i or e) stood for the sound “zh”. These sounds then merged in Spanish so that
both were pronounced “sh” which could be written with either <x> or
<j>. Another sound change then
happened which changed the “sh” sound to the velar fricative /x/. After this there was a spelling reform which
changed most of the words containing “x” (such as hixo ‘son’ and axedrez
‘chess’) so that now they were written with “j” (hijo, ajedrez). Thus Spanish went through the following
stages.
1.
<j> = /zh/, <x> = /sh/
2.
<j> and <x> = /sh/
3.
<j> and <x> = /x/
4.
<j> = /x/, and <x> not used much
Certain proper names, however, were
allowed to keep their written x’es. Like
“México”. This name comes from the old
Aztec name for the heartland of their empire, which they called Mēšiʼko. In colonial Nahuatl this name was written as
<mexico> and was, at first, only lightly hispanified into the pronunciation
/mešiko/, which sounds like “Meshico”.
When Spanish turned all of its “sh” sounds into /x/ this included the
Spanish pronunciation of “Mexico” which thus now had the pronunciation you’ve
heard that resembles “Mehico”. Note that
this sound change happened only in Spanish—it did not happen to Nahuatl, whose
speakers continued then and now to call the country Mēšiʼko. Meanwhile in English, the word was at an
early date given a spelling pronunciation of “Meksico” which is how we say it
to this day.
Texas is another example. This name comes from the Caddo word tayshaʼ
meaning ‘friend’, pronounced like /tɛyšaʔ/ ~ “TAY-sha”. The final “s” put at the end is the Spanish
plural suffix. The Caddo version of this
name ends in a glottal stop which neither Spanish nor English really have, so
this name was pronounced by the Spaniards of the early 17th century as /tešas/
~ “TAY-shahss”. Then the usual sound
changes happened, turning this into /texas/ ~ “TAY-hahss”; while once again in
English the name received a spelling pronunciation sounding like “Teksas”. What makes this different from the Mexico
example is that Spanish often does spell it with the replacement
consonant “j”—especially in derivative words like tejano.
Sometimes you need to cross-reference
the same name written in different languages in order to figure out how it
originally sounded. The extinct Jumano
tribe of Texas are now referred to as the /humano/ because that is what the
obviously Spanish-shaped spelling implies.
But in earlier Spanish they were called the “Xumana”—and the time period
makes it unsure whether this spelling was stamped according to the pre- or
post-sound change rules. In other words,
was it Humana or Shumana? The French
writings refer to this same tribe as the “Chouman”—thus we can triangulate that
the name originally began with the sound “sh” . . . Right? Well actually, in this case it’s particularly
sticky, because the early Spaniards also wrote the name of these Indians
as “Humana”. So the truth is that I really
don’t know, but it’s an interesting problem.
More archaic relics of Spanish can be
found in that semi-primordial era of north-of-the-Rio colonialism: the
1500s. The town of Waxhaw in North
Carolina is located not far from the location of an Indian settlement recorded
in the records of the Juan Pardo expedition as “Gueça”. The Spanish language having no /w/ consonant,
it was often represented as <gu> or <hu> (as in “Naguatex” from
Caddo Nawidish). But the letter ç
here is more interesting. This letter
corresponds to “z” in modern Spanish orthography—pronounced “s” in Latin
America and “th” in Europe, both of these deriving from an original
pronunciation of “ts”. Thus “Gueça” and
“Waxhaw” are plainly two different spellings of the same name, and this fact in
turn informs us that the Spanish were still using ç to write “ts” (very similar
to “ks” in the English version) and that the English name originally had
dictionary-style long ay in the first syllable (Wāxhaw)—though locals now
pronounce the name as “whack saw” [note D].
French has been a bit more stable as
compared to Spanish, at least since 1604 when it became a permanent fixture in
North America. But rewind a further
century and you’re in another realm: of the four primary sources of the De Soto
expedition, three are written in Spanish and the fourth in a variety of French
sufficiently outdated that it’s still referred to as “Middle French”. (English left its “Middle” period a century earlier,
and “Middle Spanish” isn’t really a thing.)
Other records come from the attempts by Cartier and Roberval to found a
colony on the Saint Lawrence in the 1530s.
These include the very early records of a still-not-quite identified
Iroquoian language spoken at the indigenous settlements of Hochelaga and
Stadacona. I’ll let Wallace Chafe
explain:
“The Stadaconan sounds were filtered through the phonetic
expectations of a speaker of French, a fact that led to many inevitable
distortions of what was actually heard. [... V]arious changes in pronunciation
were taking place in the French language at the time, and French orthography
had not become stabilized. For example,
for at least some French speakers the sound represented earlier by the letter s
after a vowel had been replaced by a lengthening of the vowel, but the s
was still being used as an indication of the lengthening, as in teste. Later the s was replaced by a
circumflex accent over the lengthened vowel, as in tête. This orthographic situation may help to
explain spellings like anondasco for ‘legs’ in the first vocabulary, a
spelling that makes sense only if the s marked a lengthening of the
preceding a rather than an actual s sound. The word can be reconstructed as having
sounded something like anǫdá·kǫ·.”
(Wallace
Chafe, in Decentring the Renaissance, 2001)
By the following century neither
French nor Spanish nor English had a way to distinguish quantitative length in
vowels. The Germans still do it though:
they use h and doubled consonants, as in fallen [falən] versus fahlen
[fa:lən]. In 1839 the traveler Prince
Maximilian of Wied was thus able to record long vowels in the Mandan
language—something that English speaking observers (including linguists) didn’t
reliably do until well into the 20th century.
For example, Maximilian wrote the Mandan word for turkey (máanu)
as “máhnu”, with the acute accent marking stress and the h marking vowel
length (Ryan
Kasak, A Grammar of Mandan, 2024:146).
And to finally get back to
Wahunsonacock: I cannot tell you for certain how it was pronounced, in either
17th century Powhatan or 17th century Early-Modern Virginia English. But given the spelling conventions of the
time laid out in this post, I would expect that the first two syllables were
probably more like “wah-hun” /wahʌn/ or “way-hun” /wehʌn/, rather than
“wah-hoon” as the documentary narrator said it.
A more recent example of the narrator’s mistake comes from an early 20th
century orthography of Tlingit. From
James Crippen’s dissertation:
“The NS1 orthography [of Tlingit] is characterized by a very
English-like set of vowel symbols . . . Rather infamously, the street called
‘Yandukin Drive’ by the airport in Juneau derives from a representation in the
NS1 orthography. The Tlingit term is
actually yaa ndakín /jà: ntàqʰín/ ‘it is flying along’, which in the NS1 orthography
would be either <Yandukín> or <Yah ndukín>. This spelling has unfortunately led people to
pronounce it as [jæn.ˈdʌkɪn] or [jæn.ˈdu.kɪn] in English, and it is
uninterpretable by modern Tlingit readers.”
(The Syntax
in Tlingit Verbs, p 842)
Crippen describes this in disapproving
terms, but if “Yan-dookin’ Drive” has indeed developed common currency among
the Juneauans, then of course... that is its name.
 |
| Yandukin Drive in Juneau, Alaska. This photo was reportedly taken by a mysterious man wearing black. |
Notes:
[note
A]: One of my biggest pet peeves around this
topic doesn’t come from American history at all, it comes from fiction. On the tv show Game of Thrones, the
Dothraki word “khaleesi” is always pronounced how it looks when the characters
are speaking English (or “Westerosi” I guess), but when there is a scene spoken
in Dothraki this word is pronounced /χale:si/.
When professional conlanger David Peterson was making the language for
the show, he clearly saw this word “khaleesi” and figured that “ee” must be a
quantitatively long version of /e/, as it would be in a technical orthography
devised by a linguist. But why would
you assume the narrator of A Song of Ice and Fire is using a technical
linguistic orthography for anything?
Surely the word as written in the text is just how the characters
in-universe would write it in English/Westerosi, and they’re not linguists
either. This is a shoddy job of
reverse-engineering a language. Contrast
it with the truly inspired decision by Marc Okrand to say that the “kl” in the
word “Klingon” was based on English-speakers in the Federation mishearing a
lateral affricate /tɬ/, based on real world precedents which I’ll discuss in
part 3.
[note
B]: Historians usually settle on either
“Tanacharison” or “Tanaghrisson” but I’m not sure anyone ever actually called
him by the latter. I wanted to check
which variants of his name were given by French vs English authors so I did a
little digging, but I could not find the spelling of “Tanaghrisson” anywhere in
the writings of George Croghan, Conrad Weiser, George Washington, Christopher
Gist, Marquis Duquesne, Philippe-Thomas Joncaire, William Johnson, or any of
the sources listed by the Dictionary of Canadian Biography. Nevertheless most historians treat it like
the default spelling, idk. The ones I
did find are: (French) Thaninhison, Thaninhisson, thanenhisshon, thanenhisson, Thaninhison,
Tanahisson, Tanarisson, thanenhison; (English) Thonariss, Thonarison,
Thonorison, Tanachrisson; (English written by a German) Thanayieson,
Tannghrishon, Tanacharisson, Tanacarisson.
“Tannghrishon” is probably a misprint of Tanaghrishon. The DCB lists a “Deanaghrison” and a
“Johonerissa” but I don’t know where they’re from either. I haven’t asked any specialists about this
name’s etymology. It looks like the form
might be close to /thanęhrisǫ/. Initial tha-
would be the duplicative prefix t- followed by the 3rd person singular
masculine subject prefix ha- and the final –ǫ would be the continuative aspect
suffix. –nęh- also looks like the
incorporated noun ‘corn’ (cf other regional names like Cornplanter and
Cornstalk). Also, the normal modern
Mohawk word for Bible is ohiatonhseratokénhti, but initial o- versus ka- in
Iroquoian is not very consequential. The
Seneca sound change is discussed in Charles Julian’s doctoral dissertation A
History of the Iroquoian Languages (U of Manitoba, 2010).
[note
C]: In the Hopi language both words actually have
a long o vowel (Mookwi, mooki).
Another tribe that changed its name because of an anthropologist are the
so-called “Nakota” Sioux, but the details are a bit different and anyway it’s a
story for another time. It’s also worth
pointing out that for some reason the “exonyms bad” faction seem to only apply that rule to
English, and are unbothered by Cheyenne-speakers calling the Comanche “Šé’šenovotsétaneo’o” instead of Nʉmʉnʉʉ, Ojibwe-speakers calling the Sioux “Bwaanag” instead of Dakhóta, or
Delaware-speakers calling the Iroquois “Opànuwàk” instead of Haudenosaunee,
et cetera, et cetera, et cetera.
[note
D]: Regarding the Spanish sibilants: alongside
the changes to x and j, there was a parallel shift in Spanish—of
which I know less of the details—which gave rise to the present state of Latin
American seseo versus Castillian thetheo.
Both variants descend from a state of affairs more like modern day
Basque, where “z” (of which “ç” is an older form) is a laminal /s/ and “s” is a
retracted apical /s/. This only
occasionally pops up, such as in the Nahuas calling the Castillian Spaniards
the caxtīllān because they apparently perceived 16th-century Spanish’s
apical “s” as being more similar to their own /š/ than to their /s/. Also, for the record I am not really
convinced by Frank Siebert’s Catawba etymology for Gueça/Waxhaw—which is why I
didn’t mention it—but he said it was from wátsa·ʔ (Esaw dialect) ~ wítsa·ʔ
(Saraw dialect) “crest of hair” (given in Blair Rudes, Place Names of
Cofitachequi, 2004).
Various
other etymologies are mostly from the Handbook of North American Indians,
from William Bright’s Native American Placenames of the United States,
or from Mii Dash Geget (p.c.).
.png)
.jpg)
.jpg)
.jpg)
