Counting Eskimo Words for Snow

Certainly languages tend to define reality differently - as "river" and "stream" in English define size, while "fleuve" and "riviere" in French describe whether or not the river meets the sea . Similarly do we define a buttercup as a "weed" with all its negative connotations or a flower with its positive ones. Is the criterion "wildness" or "beauty"? How might this categorisation differ in other languages - and how then might other language speakers perceive things differently?

The theory, based on the Sapir-Wharf hypothesis, and originating in the work of Boas and of Whorf, has its merits, although there is a view that it is as likely that language depends on and reflects thought as much as thought depends on language.

We must clarify:
• what is an Eskimo
• what is an Eskimo language
• what is a word
• how many references are there to snow in a given Eskimo language.

1. What is an Eskimo
Eskimos are native peoples living in arctic Canada, northern Alaska and Greenland. While the term Inuit is preferred to Eskimo by many in Canada, the term is retained here because
(a) it properly refers to any Eskimo group, not only the Inuit;
(b) its use is widespread amongst English language users for historical reasons. Details of current usage of the terms Eskimo or Intuit can be found here.

2. What is an Eskimo language
There are five Eskimo languages, of which the best-known is Inuit, spoken in a series of well-differentiated dialects ranging from Northern Alaska, across the Canadian far north, and up to the coast of Greenland.
The Eskimo language used for this article is Central Alaskan Yupik which is spoken by about 13,000 people in the coast and river areas of Southwestern Alaska from Norton Sound to Bristol Bay.
The specific dictionary used is Steven A. Jacobson's (1984) Yup'ik Eskimo dictionary.
See also words from Labrador and West Greenland

3. What is a word.
This is not as simple a question as may be commonly supposed. Woodbury suggests "lexemes" are more useful than "words."
A lexeme is like an independent dictionary entry. For example English has a single lexeme speak which gives rise to inflected forms like speaks, spoke, and spoken. Eskimo/Inuit languages have such complex inflections that each single noun lexeme may have about 280 distinct inflected forms, while each verb lexeme may have over 1000.
These languages are the prototypical example of a polysynthetic language, wherein one word contains several elements of the situation. This allows very complex ideas to be expressed in one word, e.g. 'tikitqaarminaitnigaa' "he said that he would not be able to arrive first".
Thus "my snow", "your snow", etc., would each be one word in Inuit, a stem form with a possessive affix.

4. How many references to snow
Here is a list of lexemes referring to snow in Yup'ik.
It is difficult to count because of the question of whether each lexeme can be counted separately. Woodbury offers these criteria:

(a) Do we count speak and speaks and speech as different items?

(b) Do words/lexemes with these meanings really count for you as words for snow?

(c) Do you count synonyms separately or not?

(d) If you decided to count synonyms together, will you also count together the members of noun-verb pairs which have basically the same meaning? eg swim can be a noun or a verb

(e) Some of the lexemes only occur in a small area of the Central Alaskan Yupik-speaking region. Are you going to try to make counts for each separate dialect?

5. Counting the lexemes for snow

(1) Snowflake
qanuk 'snowflake'
qanir- 'to snow'
qanunge- 'to snow' [NUN]
qanugglir- 'to snow' [NUN]

(2) Frost
kaneq 'frost'
kaner- 'be frosty/frost sth.'

(3) Fine snow/rain particles
kanevvluk 'fine snow/rain particles
kanevcir- to get fine snow/rain particles

(4) Drifting particles
natquik 'drifting snow/etc'
natqu(v)igte- 'for snow/etc. to drift along ground'

(5) Clinging particles
nevluk 'clinging debris/
nevlugte- 'have clinging debris/...'lint/snow/dirt...'

(6) Fallen snow on the ground
aniu [NS] 'snow on ground'
aniu- [NS] 'get snow on ground'
apun [NS] 'snow on ground'
qanikcaq 'snow on ground'
qanikcir- 'get snow on ground'

(7) Soft, deep fallen snow on the ground
muruaneq 'soft deep snow'

(8) Crust on fallen snow
qetrar- [NSU] 'for snow to crust'
qerretrar- [NSU] 'for snow to crust'

(9) Fresh fallen snow on the ground
nutaryuk 'fresh snow' [HBC]

(10) Fallen snow floating on water
qanisqineq 'snow floating on water'

(11) Snow bank
qengaruk 'snow bank' [Y, HBC]

(12) Snow block
utvak 'snow carved in block'

(13) Snow cornice
navcaq [NSU] 'snow cornice, snow (formation) about to collapse'
navcite- 'get caught in an avalanche'

(14) Blizzard, snowstorm
pirta 'blizzard, snowstorm'
pircir- 'to blizzard'
pirtuk 'blizzard, snowstorm'

(15) Severe blizzard
cellallir-, cellarrlir- 'to snow heavily'
pir(e)t(e)pag- 'to blizzard severely'
pirrelvag- 'to blizzard severely'

So in this list there are fifteen significant lexemes plus several varieties for each.

See also the list from Labrador and West Greenland which gives more words for snow.
Depending upon which decisions you come to about the criteria above you may come to a different total.

6. English snow lexemes (22 for comparison)

avalanche	sleet
blizzard	slush
dusting	snow bank
flurry	snow cornice
frost	snow fort
hail	snow house
hardpack	snow man
ice lens	snowball
igloo (Inuit iglu 'house')	snowflake
pingo (Inuit pingu(q) 'ice lens')	snowstorm
powder	spindrift

Do you agree with the list?
Does it conform to the same criteria as you decided for the Eskimo words?
What is your view on the Sapir/Whorf hypothesis now?

Footnote:
The dialect area abbreviations used:
NS Norton Sound dialect
NSU Norton Sound, Unaliq subdialect
HBC Hooper Bay-Chevak
Y Yukon River area subdialect of General Central Alaskan Yupik dialect
NUN Nunivak

Harry Mount in "How England Made The English" makes several claims about precise linguistic terms for landscape items. He suggests the Sami people in Arctic Scandinavia have more words for snow than do the Inuit.

Furthermore he claims that the Anglo Saxons in England had more than 40 words for "hill" depending on its exact appearance and characteristics. He points out that describing "the tall hump-backed hill by the stream" or whatever, was the best way to find your way without a map.

"So, a place ending in -dun (the North and South Downs were originally Duns) meant a flat-topped hill that was easy to build on. Those ending in -hoh, like Ivinghoe ... were sharply projecting, heel-shaped hills. Endings in -hop meant a valley or, more precisely, a remote place enclosed by hills."

Lexemes referring to snow in Steven A. Jacobson's (1984) Yup'ik Eskimo dictionary.

Based on an article by Anthony C. Woodbury University of Texas at Austin July 1991
See also "The great Eskimo vocabulary hoax" by Geoff Pullum.

See also Mark Halpern, Language & Human Nature (Transaction Publications, 2008).