Five Hundred Eskimo Words/Chapter 3

There is a possibility that the four words which are preserved, through the saga of Erik the Red, from the “Skrælings” on Markland are of Eskimo origin.^[1] If so, this is the oldest known specimen of the Eskimo language, or more acurately defined, the dialect on the Atlantic coast of Labrador. It was, however, not until the north-west expeditions of the Renaissance period that indubitable Eskimo words were written down, there being a short list of seventeen words dating from Frobisher's visit to Baffin Land in 1576. Ten years later, the first actual list of words was compiled, being of double the length as the one mentioned above, from West Greenland, and it was Davis, who in this modest way laid the foundation of the later thorough study of the language of Greenland. Thus Europe obtained, at a very early period, a superficial knowledge at any rate of two of the branches into which the East Eskimo dialects are divided, viz. the Greenland and the Central branch. As mentioned above, the very oldest information possibly dates from the third and last, the Labrador branch.

This geographical division into a Greenlandic, a Central and a Labrador branch of dialects, is practical, but probably of no great linguistic value. For irrespective of these branches the dialects form groups, the mutual relation of which only in part corresponds with the geographical division. The West Greenland group is rather close to the Iglulik group, which belongs to the Central branch, the latter in its turn being very close to the Labrador group. On the other hand, the difference, for instance between the Pâdlimiut and the Iglulik dialects, is far greater than between the Iglulik and the Labrador dialects.

One thing must be borne in mind which, however, will fully appear from the list, viz. that even between the dialects mutually ​most removed the differences are only small. It is true that they are somewhat greater than between English and American-English or between Danish and Norwegian "riksmål", but they are less than between Danish and Swedish. The difference between the individual groups may, perhaps, be compared with the difference between standard English and Scotch, and between standard Danish and Jutlandic. A similar amount of practice to that which is needed by an Englishman in order to read Robert Burns, or by a Dane in order to enjoy Blicher, is all a Greenlander requires in order to understand the central dialects. Neither Jacob Olsen, a native of West Greenland, nor our Polar Eskimos had spent many days among the Canadian Eskimos, before they were able to talk freely with them upon all everyday subjects.

Within the confines of Greenland there are three groups of dialects which belong, respectively, to the east coast, the west coast south of Melville Bay, and the Thule or Cape York District at Smith Sound.

The most easterly of these — in fact the most easterly of all the Eskimo groups — only comprises the southern part of the east coast of Greenland, from Blosseville Coast to Cape Farewell. It is true that as late as at the beginning of the 19th century the Eskimo language was spoken much farther north, viz. on Clavering Island, between lat. 74–75° N, where in 1823 Clavering met a family of twelve members, including women and children. However, the inhabitants of the north-east coast were entirely extinct, before the next expedition visited these remote parts, and nothing whatsoever is known of their language. Thalbitzer defines the distribution of the three southern dialects in the following manner:^[2]

Lat. 68°–64° N.	Kialineq (Kialêq)-Pikiutdleq (Pikîteq): the Angmagssalik District.
Lat.„ 64°–62° N.	Umîvik-Ikermiut (Ikermîn)	${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\\\ \ \end{matrix}}\right\}\,}}$	Frederik VI Coast.
Lat.„ 62°–60° N.	Anoritôq (Anoritêq)—Aluk (Alik)

The only vocabulary to hand from the east coast was for a long time the brief list, published by Rink, and based upon the notes made by the Holm Expedition. Since then the Angmagssalik dialect has been subjected to exhaustive studies by Thalbitzer, whose results, however, have only been published in parts. In 1923 ​it was spoken by 709 individuals. From Frederik VI Coast very little is known apart from a list of words and a text published by the Rev. Poul Vibæk, and as the inhabitants of the whole of this region have gradually immigrated to the southern west coast in the course of the 19th century, their linguistic peculiarities have probably now entirely, or almost entirely, disappeared. As the East Greenlands dialects are not included in the present work, no particulars regarding them will be given. —

On the whole of the long west coast which, irrespective of the indentations, corresponds with the distances between Copenhagen and Naples or from Niagara halfway into Florida, four principal dialects are spoken, which, however, are very closely related. Even the oldest vocabulary from these parts, viz. the above-mentioned work of Davis, brings out a point of considerable interest. Davis translates "a fogge" by tucktodo,^[3] whereas "fog" in West Greenland is now called [pujɔq]. At the present time it is not till Iglulik that the word [taktuɳ] is again met with — at Kazan River [takxᴇq] — with the same meaning. The word is derived from [tᴀ·q], darkness.

Whereas Davis' list was written down in far too inadequate a manner to be of value from a phonetic point of view, this, in spite of all deficiencies, does not apply to the two succeding compilations, viz. those made by Olearius and Caspar Bartholin, both of them originating from the Greenlanders who in 1654 were carried from Godthaab Fjord to Denmark by the Danell Expedition. These lists and, in a much higher degree, the later, far more complete dictionaries compiled by Poul Egede and Fabricius fully show that the pronunciation in West Greenland has altered on certain points during the last centuries. The interesting feature is that the old forms show a far greater resemblence to the Central Eskimo pronunciation than do those of the present day. In modern West Greenlandic an assimilation has taken place in the first sounds within certain consonant groups, whereas in former times they were evidently pronounced separately, as is still the case in the Thule District and within the Central Eskimo area. Instead of the modern forms [umiʃ·uit], large beard, [iʟ·ät·ut], those (who) laugh, [tut·o], caribou, Olearius and Bartholin give the forms umixuin, iglakton, tucto, which clearly indicate the pronunciation of the consonants.^[4] Even a comparatively modern author like Fabricius writes the words in a manner which points in the same direction, ​for instance, túkto.^[5] The writing of a mute g (corresponding to the k and c of older authors) has been retained in the Greenlandic orthography. There is evidently some connection between the assimilation of the initial consonant and the change from voiced to voiceless which seems to have taken place in these cases, as for instance ublau, day, now [uʟ·ɔq]; ibling, thou, now [iʟ·it]; ipgin, straw, now [if·it].^[6] Fabricius has these words as údlok (but kablunâk, a European) iblit or ivlit, as well as igvit.^[7]

An assimilation of consonants which was demonstrated by Thalbitzer^[8] is the disappearance of r before s, for instance [iʃe], eye, written irse by Egede.

Another change applies to the final consonants which in the 17th century seem to have been nasalized in a much higher degree than at present. Instead of the present q, k and t, the old lists have spellings which suggest the pronunciation q̃, ɳ and n. Here a few examples:

Olearius adds: “Es sind viel Wörter / sonderlich die das (n) bey sich haben / die sie durch die Nase reden / und nicht so wohl können geschrieben werden".^[10] According to Hans Egede the nasal final consonants particularly occurred among the women.^[11] The old authors also have m as a relative ending instead of the modern p. Thus, Hans Egede writes Nallekam okausia, the Lord's prayer, whereas it is now [na·läk·äp ɔqa·sia]. As early as in Fabricius' works these nasals are replaced throughout by the modern tenues, and at the present time they only occur in certain instances where the final consonant is affected by the following word, for instance [kikun·uko], ​who are they, compared with [kikut], who, with a normal plural. Also, according to Jacob Olsen, the Greenlanders round Sukkertoppen to a very great extent still use ɳ, n and m as final consonants instead of k, t and p.

The words quoted above show a further difference between older and modern West Greenlandic, viz. as regards voiced and voiceless consonants: [qᴇ̃r̃nᴇʀtɔq̃], [qᴇ·ʀʟutɔq̃] with the voiceless ʀ, where in modern times r is used. The same pronunciation is shown by a form pilachtagtong [piläxtᴀʀtɔq̃], he (who) flenses,^[12] as contrasted with the modern [pilät·ᴀrtɔq]. Here Fabricius has kernektok, kærtlutôk and pilléktorpok.^[13] Bartholin's ochac, blubber, even reminds us of the Pâdlimiut form [ɔqʀɔq], Fabricius orʃok, modern West Greenlandic [ɔrʃɔq].^[14]

Finally, the disappearance of the diphthongs should be mentioned as the last change in the pronunciation. However, in West Greenland the diphthongs au and ai (ao and ae) are still written, while the younger generation now only pronounce a long a; even Thalbitzer in his work from the journey of 1900–01 writes these diphthongs as a" and a', and in some localities at any rate, for instance at Jakobshavn, they seem within living memory to have been pronounced still more distinctly.

The modern West Greenland dialects, which are spoken by 13,847 individuals all told (1923), have particularly been studied by Thalbitzer, and while referring the reader to the work of this author I shall limit myself to a few remarks. We are able to distinguish between at least four dialects, although it is rather difficult to define them closely. There are gradual transitions, as some peculiarities of pronunciation seem to have spread beyond the confines of the dialects, where they must rather be supposed to belong. As I think I have proved on an earlier occasion,^[15] the southern dialect, at the time of Hans Egede, did not extend beyond Julianehaab Bay, while its influence is now to be traced all the way to Godthaab. This is chiefly due to the Moravian missionaries who induced several southerners to settle near Godthaab, when they passed through this district on their trading journeys. The four main dialects roughly comprise the following areas:

​The Julianehaab dialect shows a connection with the language of the east coast in several respects, not merely in the vocabulary, but also phonetically. The most striking feature is that the vowels i, e, ᴇ in many cases, as further east, replace the rounded u, o, ɔ, for instance [mᴀrʟik], two, [ajiɳ·ilᴀq], good. In the most southerly part of the district ɳ is frequently used instead of g. The latter sound has perhaps penetrated from the north, where ɳ is unknown in the corresponding combinations.

The dialect spoken north of this district, and which comprises the west coast as far as the Arctic circle, has formed the foundation of the written Greenlandic language. As contrasted with the Julianehaab dialect it is characterized by the rounded vowels u, o, ɔ as well as by g, instead of ɳ. As contrasted with that of the Egedesminde-Umánaq area the sound-combination tˢ is common. Instead of a peculiar sound occurs which, from the specimens given to me by Jacob Olsen, must rather be characterized as a cacuminal point-s, which harmonizes with the observation of Thalbitzer.^[16] As already mentioned, the nasal final consonants ɳ, n, m are said to be retained in the Sukkertoppen District.

In the Egedesminde-Ümánaq dialect ɳ is extremely common instead of g. The latter sound is — as far as my experience goes — almost unknown in Egedesminde, but sometimes occurs at Disko Bay and in the Ūmánaq District. In addition a long t is frequently used instead of tˢ, and rɳ instead of rn. In the Ũmánaq District q̃ sometimes replaces r, which is probably owing to influence from the north. It is the Ritenbenk dialect which forms the base of the present list.

In respect to the Upernivik dialect Thalbitzer has undertaken a thorough research, with the result that it has been proved to lack the common sounds ʀ, x, g, f. Instead of the long voiceless fricatives q·, k·, p· are used, and r, v, tˢ are frequently replaced by q̃, ɳ, s and sometimes u by i. —

The third and last group of dialects in Greenland only comprises a single dialect, spoked by the. Polar Eskimos in the Thule District, a small tribe of 251 individuals (1923), who inhabit the country from the northern part of Melville Bay to Humboldt Glacier. Their dialect is very little known, although a short and, it is true, rather incorrect list of words was given as early as by John Ross, the discoverer of the tribe. A mistake which occurs throughout this list is the placing of h before words beginning with a vowel as for instance hemuck [imᴇq], water, haphuck [ᴀrfᴇq], bowhead whale. Since then Kroeber, in his description of ​the ethnography of these Eskimos, has given a number of words, his list of conjurors' terms being particularly valuable. There is the peculiarity about Kroeber's list that it seems to show a strong influence from Baffin Land. This may possibly be due to the influence of Boas' famous work "The Central Eskimos", but may also be due to the immigration which took place from Baffin Land to the Thule District in the sixties of last century. The geologist Stein in his treatise on geographical names among the Polar Eskimos gives a number of phonetical data, but he also makes remarks which suggest an influence from the other side of Baffin Bay (the presenze of z and q̃, the frequent use of nasal final consonants, etc.).

The vocabulary of the Thule dialect deviates in many ways from that of West Greenland, and phonetically this dialect represents a more old-fashioned stage. The far-reaching assimilation of consonantal groups and diphthongs has not yet taken place, and this evidently has some bearing upon the preservation of the voiced consonants b, d, g, for instance [täblo], chin; [udlᴀ·q], morning; [iglᴇq], sleeping platform. Other voiced sounds, which occur in West Greenland, are here sometimes used in different combinations, thus [sᴇrva], guillemot; [tiggäk], stinking fjord seal.

It is of the greatest interest to find faint traces in this dialect of the course of the metathesis, which Thalbitzer demonstrated among the East Greenlanders, forms like [upᴇ^rɳgᴀ·q], spring, [ᴀ̃·^rɳguᴀq], amulet, being met with. On the other hand, the rdl-forms are already fully developed: [mᴀrdluk], two; [pᴇrdlᴀ·q], sledge shoe; the same applies to [mᴇrqut], needle. The corresponding retrogressive labialization, which has taken place in West Greenland, but not in Labrador, is also found at Thule, for instance [iblit], thou. This is a sound-development which closely connects the Thule dialect with the other Greenland dialects and distinguishes it from the Central Eskimo and Labrador ones.

Also as regards the s·sounds, the dialect seems to be most closely related to West Greenlandic. The dental s, however, rarely occurs, but in most cases it is replaced by ʃ or h. After k the latter sound is pronounced with the back part of the tongue so greatly raised that it becomes x, for instance [ikpakxᴀq], yesterday. On the other hand j is presumably not used instead of s, as believed by Thalbitzer.^[17] Nasal final consonants occur, but they are not so common as the corresponding tenues. In the above I have suggested a possible explanation of their frequent use in earlier ​works. It is hardly the language which has changed, for one of my native informants must have been about sixteen years of age, when Stein wrote his treatise.

One thing which makes the investigation of the Thule dialect so extremely difficult is the slurred, vaguely articulated manner of speaking, by which a number of sounds, especially the final consonants, are sometimes entirely swallowed. Thus in daily speech [qᴇqᴇrtᴀrhuk], island, sounds rather like [qᴇq^ᴇrtᴀ^rhu^k]. A q at the end of a word is not infrequently weakened to r, while r between two vowels is labialized to resemble w.

It is apparently possible to divide the Central Eskimo dialects into five groups, which may be termed the Cumberland Sound, the Iglulik, the Netsilik, the Coronation Gulf, and the Barren Grounds groups.^[18] Like the Thule dialect, they are all characterized by the non-assimilated consonant-groups and diphthongs, but otherwise the mutual differences between them are rather considerable.

The Cumberland Sound group belongs to southern Baffin Land and is spoken by the Eskimos inhabiting the northern coast of Hudson Strait, Frobisher Bay, Cumberland Sound and Home Bay. Frobisher's list of words, which is the oldest incontestable information relating to the history of the Eskimo language, originates from these tribes, and as pointed out by Thalbitzer even this early list is of considerable interest from a phonetic point of view. Owing to Boas' vocabulary and texts, the Oqomiut dialect at Cumberland Sound has long been the best known Central Eskimo dialect. However, we are not going to enter into the details of this dialect, as the whole group falls outside the sphere of activity of the Fifth Thule Expedition. —

To the Iglulik group belong three different tribes, the Tununermiut, the Iglulingmiut [iglu'liɳmiut] and the Aivilingmiut [aivi'liɳmiut] which together, according to the census taken by Therkel Mathiassen, amount to 499 individuals. The Tununermiut have their home at Ponds Inlet, from where they sometimes extend their hunting area both to North Devon and Admiralty Inlet.^[19] It is from this tribe that an immigration into the Thule District took place in the sixties of last century. The Iglulingmiut inhabit the region ​round Iglulik at Fury and Hecla Strait, as well as the whole east coast of Melville Peninsula, as far as Lyon Inlet. Of the tribes belonging to this group they are the least influenced by civilization. The Aivilingmiut belong round Repulse Bay and on the west coast of Roe's Welcome, as far as Chesterfield Inlet. After the extinction of the Sadlermiut, the original inhabitants of Southampton Island, part of the tribe now frequently live in this great island.

As the ethnographer must mainly turn to the expedition of Parry and Lyon for information about these Eskimos, so our previous knowledge of their language is based upon Parry's list which, in fact, is by far the best vocabulary from all Central Eskimos west of Hudson Bay. Within more recent times Boas has published a series of texts from those parts.

Within this group the Iglulik dialect has been selected as a type. It contains the two characteristic voiced consonants z and ʒ which do not occur in Greenlandic. As was the case among the Oqomiut, the nasal final consonants are far more common than tenues, which however also occur; on the other hand q is not changed intervocally as in the Oqomiut words ixadlung [iʀätʟuɳ?], trout; ixiqoq [ɪʀɪqɔq?], little finger.^[20]

As mentioned above, a retrogressive uvularization as well as labialization has taken place in Greenland, and so it will be interesting to see, how far this applies to the Iglulik dialect. A few examples will make this apparent.

It will thus be seen that like the Labrador group the Iglulik dialect has undergone uvularization, but not labialization. However, the Iglulik dialect is on a more primitive stage than that of Labrador. Thus, the rn-combinations still show distinct traces of the metathesis, and this also applies to the rq-complex. As compared with West Greenlandic the assimilation of rʟ to ʀ· and r· represent a more advanced stage which, at any rate ​in the case of the former feature, recurs in Labrador, whereas the other peculiarities are more primitive.^[21] From certain spellings which Parry uses, it may be inferred that the metathesis within the Iglulik group is of a comparatively recent date; thus he writes mādlĕrǒke, opēn-řă, il-weet or ig-weet.^[22]

Another old time characteristic, which is again met with in Labrador, is the retension of different sounds which have all been merged into s or ʃ in Greenland, for instance:

The Aivilik dialect is extremely close to that of Iglulik, and very naturally, indeed, as mixed marriages are so common that the pedigree of practically all of these Eskimos, at most a couple of generations back, includes some individual of the other tribe. Some difference is, however, to be traced in the language. The nasal final consonants are a little less common with the Aivilingmiut and, as it seems, particularly in place names, which suggests that these sounds did not originally belong in this region. Besides, with the Aivilingmiut the Iglulik d is replaced by b or g, for instance:

Another word deserves to be mentioned: Iglulik [tisamän], four; Aivilik [sitamän]. The Aivilingmiut have a marked sing-song accent, and the last syllable is frequently long-drawn.

Generally speaking, from a phonetic point of view the Iglulik group must be said to be very close both to Greenlandic and the Labrador group, though considerably closer to the latter, but also on a still more primitive stage. —

The home of the Netsilik group is, roughly speaking, the regions round the Magnetic Pole, from Committee Bay to the east ​towards Simpson Strait to the west. They constitute the smallest Central Eskimo group, numbering, according to the census taken by Knud Rasmussen, 423 individuals in all, and they are furthermore of all living Eskimo tribes those who are least influenced by civilization. While there are several sub-tribes or local groups, the following may be regarded as the main divisions: Arviligjuarmiut [arviligʒuˡᴀrmiut]; Netsilingmiut [näciˡliɳmiut]; Ilivilermiut [iliviˡlᴇrmiut]; Utkuhigjalingmiut [utkuhigʒaˡliɳmiut]; and Háningajormiut [han·iɳaˡjɔrmiut]. The Sinimiut tribe which Rae met in Committee Bay is now extinct, with the exception of two men who live in the Repulse Bay District.

With regard to the language of all of these tribes no information has been available up to the present.^[23]

The Arviligjuarmiut have their proper hunting fields on both sides of Simpson Peninsula in Committee and Pelly Bay, but from here several families have, during later years, immigrated to the regions round Repulse Bay. They have, however, not been absorbed into the Aivilik tribe, who regard the Arviligjuarmiut and the other more westerly tribes as a kind of uncivilized louts, although in their native place at any rate the latter are far abler and more well to do than the Aivilingmiut. The Arviligjuarmiut families therefore to a certain extent keep to themselves, although mixed dwelling places may occur, and they also retain their own habits and language even to the second and perhaps the third generation. Further, many Arviligjuarmiut come to Repulse Bay on trading expeditions, spend a year or two there and then return. It is this tribe whose dialect has been taken as a type. The Netsilingmiut are the tribe living on both sides of the Boothia Isthmus round the Magnetic Pole, whereas a subdivision of this tribe, the so-called Arvertôrmiut, live as far north as Bellot Strait and another, the Kûngmiut, round Murchison River. The Ilivilermiut, better known under the less correct name of Ugjulingmiut, constitute the population at Simpson Strait. The Utkuhigjalingmiut, who formerly lived at the mouth of Back River, have now, through the influence of the Hudson's Bay Company's trading station at Baker Lake, become a perfect inland tribe, hunting caribou and trapping in the lower reaches of Back River south of Lake Franklin. The Háningajormiut and Ualiardlît, who from of old are inland tribes, now ​only consist of a few families round the great lakes at the middle course of the same river.

In this group we are, for the first time, confronted with the peculiar sound, the "glottal stop" which has hitherto not been recorded from the Eskimo language. It occurs in words like [i'nᴇq], fire, pyrites; [qi'mᴇq], dog; [u'läktɔq], he runs, in the Iglulik dialect [inɳᴇq̃], [qiɳmɪq̃], [udläktɔq] respectively. Here the Netsilik and the Iglulik groups seem to have taken different courses. Thus in the forms corresponding to ignerk (Mackenzie Delta), knak (South Alaska), eknek (Asiatic Eskimos),^[24] the Netsilik-group has weakened the consonant to a glottal stop, whereas the Igluliks have nasalized it.

The position of the Netsilik dialect as regards metathesis appears from the following examples, which should be compared with the words on p. 24.

The two first and the two last-mentioned Netsilik words in this list resemble the Greenlandic having l as a last consonant, combined with r and b, respectively. This is explained by the sub-joined West Eskimo forms: they are the same words which at Mackenzie River and in Alaska have these combinations of sounds, and so ​they cannot be classed with the words which have been subjected to metathesis. On the other hand, all the other words are destinguished from the Greenlandic, or, in other words, the retrogressive uvularization and retrogressive labialization have not taken place within the Netsilik group, any more than among the Western Eskimos. In this connection I should like to call attention to two other groups of consonants where also in some cases metathesis has possibly taken place, viz. old Greenlandic ks, now assimilated to ʃ· or s·, and Greenlandic ɳm, now assimilated to m·. The former complex occurs in the rare word [qaʃ·e], house of assembly or dancing house, which was spelled kakse by Fabricius. The Netsilik tribes as well as the Caribou Eskimos pronounce this word [qaç·e] with a long, faintly palatalized g; but sometimes the form approaches [qaᶻge]. Barnum writes kazhga as occurring in South Alaska. It is a likely supposition that the Alaskan form is the oldest. Unlike in Greenland, no metathesis has taken place with the Central tribes, but on the other hand there has been an assimilation of the two sounds to one palatalized sound. As to the nm-complex I have only the Cape York form [qaɳmäk], ankle, to support my hypothesis, the pronunciation on Simpson Peninsula as well as at Kazan River sounding like [qamɳᴀq]. Both in West Greenland and in Alaska the words for ankle are derived from another root, and so cannot be taken into consideration. As with the rl- and bl-groups there are, however, cases where ɳm must be original and cannot be due to any metathesis; this, for instance, appears from [kiɳmiᴀq], mouth-piece for bow-drill (Thule District); [kiɳmiᴀq] (Simpson Pen.); kinmiak (Mackenzie R.); kengméak (North Alaska).^[26]

To the west of the Netsilik tribes, at Bathurst Inlet and Coronation Gulf, we find the fourth Central Eskimo group of dialects, viz. the so-called Copper Eskimos. The language has been thoroughly investigated by Jenness, a member of the Canadian Arctic Expedition, and therefore it will be more correct to postpone a discussion of its peculiarities, until the results of this author have been published. —

The Barren Grounds group is the most southerly of all Central Eskimos. It comprises the tribes of the Qaernermiut [qa^ᴇrnᴇrmiut], the Hauneqtôrmiut [haunᴇqtɔ·rmiut], the Harvaqtôrmiut [hᴀrvᴀqtɔ·rmiut], and the Pâdlimiut [pa·dlimiut]^[27], who in all amounts to ​about 500 individuals. The Qaernermiut is the tribe which was formerly known under the name of Kenipitu, a whalers' term, which was never used by the Eskimos themselves.^[28] They live round Chesterfield Inlet and Baker Lake. The Hauneqtôrmiut tribe, which only numbers a dozen families, hunt caribou within the coast between Rankin and Dawson Inlet, but come down to the sea in spring in order to catch seals on the ice. The Harvaqtôrmiut are entirely an inland tribe, their dwelling places being situated along the west-east stretch of lower Kazan River, the Eskimo Harvaqtôq.^[29] South of the two latter tribes, right to the timber line, live the largest tribe of this group, the Pâdlimiut. In spring a small number of them come down to the coast at Eskimo Point and farther south, whereas the remainder stay inland all the year round.

Although the whole of this Eskimo group has been practically unknown from an ethnographical point of view, until the Fifth Thule Expedition, there is curiously enough no other Central group, from which so many lists of words are available. Besides those of Dobbs, Gilder, Tyrrell and Hanbury, we have the words communicated by Father Gasté in the dictionary of Petitot. However, they are all of comparatively little value, as the words are frequently written down in such a manner that unless one knew them beforehand, it would hardly be possible to understand them.

The oldest list is that of Dobbs, which goes right back to the middle of the 18th century. He does not mention from where he has his some hundred and fifty words and short sentences, but they are clearly from a tribe of the group dealt with here. Spellings like petick-sic, bow, se-la-luk-toke, it rains, sha-veck, iron,^[30] suggest that they do not date from the Pâdlimiut, for they have now, at any rate, only h. This sound, however, Dobbs has heard in ehick, eye or rather eyes (dual.). In all probability a shifting of the tribes has taken place after the time of Dobbs,^[31] which makes it impossible to say anything with certainty. Neither does Gilder tell us much about his list of words, beyond the fact that he got "much information" from a Qaernermio.^[32] His list seems to ​be rather unmixed, though not entirely Qaernermiut, for a word like mok'-uk-too, it rains, decidely appears to have come from Aivilik. Further, the words of Gasté are beyond a doubt from the Pâdlimiut, which also appears from the introduction written by Father Petitot.^[33] Owing to his nationality, the compiler of this list has failed to hear the h'es, for which reason he writes, for instance naak [nahᴀq], fur hood, anelpot [hanälrut], whittling-knife, ik-ik [hikhik, hikxik], marmot. Tyrrell's list is the only one originating from the inland, his informant being a native of the regions round upper Kazan River. Hanbury mainly collected his material in Churchill from the old Hudson's Bay Company interpreter, Geo. Oman, a half-breed Indian, and thus the list is rather Pâdlimiut, but naturally without any phonetic value. A number of words, however, are indicated as originating from Baker Lake, or in other words from the Qaernermiut. The present list is also Pâdlimiut, compiled in a dwelling place at the outlet of Kazan River from Lake Hikoligjuaq, but the words associated with the sea have been supplemented from the coast group at Eskimo Point.

There is no reason to enter in detail into the phonetics of the Pâdlimiut dialect, as it shows rather close agreement with the Netsilik group. A glance here and there at the list of words will persuade us that retrogressive uvularization and labialization have not taken place. As mentioned above the s and ʃ of the Netsilik dialects are everywhere replaced by h, and when drawing conclusions from analogous cases in other languages, it seems natural to considers to be the primary sound. Neither are the nasal final consonants heard. Glottal stops are still more marked than in the Netsilik dialect, so that the Pâdlimiut almost jerk and hack out their sentences, alternately with a high and a low accentuation of the syllables. Their vowels seem a little purer than the Greenland ones, a is more open, o more closed and the nasalization is inconsiderable. The Harvaqtôrmiut dialect is almost like that of the Pâdlimiut, whereas the Qaernermiut and Hauneqtôrmiut dialects are little more removed from it, having ʃ instead of h and making a less frequent use of the glottal stop.

From the preceding there can be no doubt that, within the dialects dealt with, the most primitive phonetical stage represented is that west of Hudson Bay. Neither in the region round the Magnetic Pole, nor on the Barren Grounds have retrogressive uvularization and labialization taken place. In the development of language the Iglulik tribes have followed the same courses as those of Labrador. The retrogressive uvularization seems to be oldest in the rʟ-complex, in so far it is quite correct that it is not only fully completed at Iglulik, but that, as in Labrador, a further development has taken place, by which the L has become assimilated by the r.^[34] On the other hand, the rn- and rq-combinations show their more recent origin by the fact that the metathesis has in neither case been entirely completed at Iglulik, and the same applies to the rn-complex of the Thule District, in which perhaps the metathesis is thus most recent.

Retrogressive labialization has not taken place outside Greenland. However, there seem to be certain facts relating to the complex which has been influenced by the Greenland labialization (vʟ) which, nevertheless, refer the Central Eskimo^[35] dialects now spoken to a somewhat more recent stage than the West Eskimo ones. This also applies to the assimilation of zg to g. But in both cases the material is too limited to permit of any definite conclusions being drawn.

Of other, presumably primitive features in the Netsilik and Pâdlimiut group may be mentioned the occurrence of r and ʀ in certain combinations, where West Greenland has s or ʃ, and the Labrador and the Iglulik group ʃ or j. With the Netsiliks [ugʒuk], barbed seal, and [qugʒuk], swan, frequently sound almost like [ugruk] and [qugruk]. The Pâdlimiut always have [ᴀqʀᴀq], aurora borealis, and [ɔqʀɔq], blubber. This quite agrees with Alaskan forms. Thus the dialects spoken in Alaska cannot lay claim to being treated as something apart to the same degree as formerly, although on certain points they still represent the most primitive stage. Neither should it be forgotten that in other respects than the metathesis mentioned above, they are the most advanced, viz. by dropping i in unstressed syllables and the intervocalic n and ɳ. Much still remains to be done, before linguists will be able to give an undisputable answer to the question as to where the Eskimo language was first spoken.

​Only one more thing should be mentioned. Perhaps one of the most important results of the Fifth Thule Expedition is that it was able to demonstrate that in former times a culture prevailed within the central regions, which was extremely close to the Alaska and Greenland culture, but that this homogeneous culture was split up into two parts by the subsequent penetration of inland tribes.^[36] The question then presents itself whether this fact has left any trace on the language. At any rate I believe myself to be entitled to mention one such phonetic trace, viz. the distribution of the peculiar voiceless lateral ʟ. It seldom occurs in the Central dialects, which mainly have a common voiced l, but it plays a great part in the far east of Greenland, as well as in the far west of Alaska, indeed, even in the extreme link of the chain of Eskimo tribes, the Ugalagmiut. As stated by Buschmann,^[37] it is a sound with a marked western distribution in North America, as it also occurs among the Tlingit, the Athapascans, the Nootka and the Nahua, and it does not seem improbable that this sound has been adopted into the Eskimo language in Alaska, whose culture has generally been subject to very strong influences from the North Pacific coast.