Grammatical Gender in Algonquian

IVES GODDARD
Smithsonian Institution

For nearly 400 years linguists and others have been trying to describe and
explain the t w o grammatical genders that pervade the g r a m m a r of the
Algonquian languages. A representative selection of statements they have
m a d e o n the subject is given in the appendix to this paper; these will be
referred to in w h a t follows b y the n u m b e r they have there.

THE HISTORY OF THE PROBLEM

The basic facts and the nature of the problem they present are evident in
Friderik Baraga's b e n c h m a r k mid-nineteenth-century O j i b w a grammar:*
Animate substantives are called those which denote beings and things
that are living, or have been living, really or by acceptation.
Inanimate substantives are called those which signify things that have
never lived.
... [Sjubstantives which signifies [sic] things that have no life at all,
but which the Indians treat in their language like substantives that sig-
nify living beings, create one of the greatest difficulties and peculiari-
ties of this language; because there is no rule by which you could be
guided to k n o w these substantives. ...
Here are some of those substantives which signify things that have no
life, but are employed by the Indians like substantives that signify liv-
ing beings:

mitig 'a tree' (nisdkosi) 'a corn-ear'

bakwezhigan 'bread' mazaan 'a nettle'
asin 'a stone' ziibwaagan 'corn-stalk'
mishiimin 'an apple' ningidig 'my knee'

1. The italics, glosses, and order of presentation are Baraga's (reading down his left col-
umn on both pages, and then down the right column); I have added explanations in brack-
ets. The transcription of the forms, however, has been converted into the standard
orthography of Nichols and Nyholm (1995), except for the word for 'a corn-ear', which is
not known today (John D. Nichols, personal communication 2001).

Papers of the 33rd Algonquian Conference, ed. H.C. Wolfart

(Winnipeg: University of Manitoba, 2002), pp. 195-231.
196 IVES GODDARD

bingwi 'ashes' agig 'cold, phlegm'

asemaa 'tobacco' diba 'igiiziswaan 'watch, clock'
akik 'a kettle' giizis 'sun, moon, month'
opin 'a potatoe' miigwan 'feather, quill'
bigiw 'pitch' ['quill' = 'feather pen']
mikwam 'ice' nabagisag 'a board'
goon 'snow' waabaabigan 'lime' [the substance]
daashki(gi)boojigan saw-mill' opwaagan 'pipe'
jiibayaatig 'cross' zhooniyaa 'silver, money'
mandaamin 'com' asab 'a net'
waabigan 'clay' es 'shell' [of mussel, etc.]
zenibaanh 'silk, ribbon' gishkibidaagan 'a tobacco-pouch'
mazinichigan 'image' miskodiisimin 'bean'
giizhik 'cedar' zhingob 'a fir-tree'
moshwe 'handkerchief zhingwaak 'pine-tree'
zhooniyaans 'shilling' minjikaawan 'a mitten, a glove'
miinensagaawanzh 'thorn' odaabaan 'a sledge'
[i.e., 'hawthorn tree'] bootaagan 'stamp, stamper'
ozaawaabaan 'gall, bile' [i.e., 'pestle']
anang 'a star' nindiniigan 'my shoulder-blade'
animikii 'thunder' miskwimin 'raspberry'
ishkodekaan 'fire-steel' bagaanaak 'a walnut-tree'
gijipizon 'belt' ozhaashaakon 'tripe de roche'
didibise-odaabaan 'waggon, cart' [i.e., Umbilicaria muehlenbergii]
godawaan 'a block' baapaagimaak 'ash-tree'
[i.e., 'large fire-log'] [i.e., 'white ash']

A n d a vast number of others. (Baraga 1850:18-20.)

T h e two genders are conventionally called b y the s a m e terms Baraga

used, animate and inanimate. H u m a n s and all other m e m b e r s of the ani-
mal kingdom (living and dead), as well as spirits, gods, and the like, fall
into the animate gender. But also in the animate gender are, as Baraga
says for Ojibwa, a "vast n u m b e r " of miscellaneous things that are not liv-
ing, and are not believed to be living. Baraga refers to these as living
"really or by acceptation"; as "substantives [for] things that have no life at
all, but which the Indians treat in their language like substantives that sig-
nify living beings" [emphasis added]; and he says, "there is no rule" to
define them. Examples include: ashes, pitch, ice and snow, ribbon, hand-
kerchief, belt, large fire-log, phlegm, glove, bile, pestle, various body
parts.
 G R A M M A T I C A L G E N D E R IN A L G O N Q U I A N 197

The terms animate and inanimate (in French, anime and inanime)
werefirstused in the veryfirstdescription of the two genders, that by
Paul LeJeune in 1634 (App.: 1). In English they werefirstemployed by
John Eliot in his Massachusett grammar of 1666 (App.:2), but his usage
was probably not independent. W e m a y reasonably speculate that these
terms were mentioned in the the long conversation about Algonquian
grammar that Eliot had with the Jesuit linguist Gabriel Druillettes in Rox-
bury in 1650 (Thwaites 1896-1901, 36:90; Powicke 1931:55; Goddard
1996b:21).
The existence of unpredictable animate nouns was recognized from
the beginning, and for three centuries their treatment as animates was
taken as strictly a matter of grammar. LeJeune did not say that animate
nouns referred to things believed by the Indians to have souls; he said that
they were things without souls that in verbal agreement were treated the
same as things with souls (App.:l). Eliot merely says that he knows
"some few exceptions" (App.:2). N o one claimed that Indians thought
everything designated by a grammatically animate noun really was alive.
In fact, early francophone linguists eventually came to call the genders
noble 'noble, of noble rank' and ignoble 'base, of non-noble rank' (App.:
3,4, and 5). Even Schoolcraft, w h o begins with talk of "the peculiar opin-
ions and superstitions of the Indians," eventually makes it clear that he
considers the animate class to be a category of the language, used for
things that are alive (or are "personified" and believed to be alive) and
also for things that are merely esteemed, held in high regard, respected, or
useful (App.: 6). Similarly, William Jones, a student of Franz Boas and a
native semi-speaker of Fox (Meskwaki), wrote that this "rigid classifica-
tion" was "notrigidlymaintained," since inanimates could be "personi-
fied" and become animate. But he added that not all grammatically
animate but lifeless objects could be explained by personification
(App.: 8).
Then, after 300 years of agreement among observers of Algonquian
gender, an entirely n e w interpretation appeared in the mid-20th century.
Some ethnologists began to insist that the nouns of animate gender form a
"culturally constituted cognitive 'set'" (Hallowell; App.: 9) or a "semantic
domain" (Darnell and Vanek; App.: 11) and are "definable" (Straus and
Brightman; App.: 13), generally as having power or the like. Useful in
making this point was Hockett's claim (App.: 10) that inanimates could
198 IVES G O D D A R D

shift to animate gender w h e n ascribed the powers of people, such as

speaking and being spoken to, or of animals and spirits, but that animates
never shifted to inanimates. Hockett suggested that this fact might
account for the "scattering" of unexpectedly animate nouns; Black-Rog-
ers wrote similarly (App.: 12). The obvious difficulty of explaining the, in
fact, rather large number of unexpected animates in this w a y has gener-
ally been slighted in this literature, though the existence of unexplained
cases is sometimes acknowledged. Straus and Brightman (1982) provide
the most extensive examination of actual data and m a k e the greatest
efforts to offer explanations of individual words, though these often
involve analogies, linkages, and special pleading ("necessarily ad hoc,"
they concede [Straus and Brightman 1982:126]), rather than the system-
atic evaluation of all nouns with respect to what characterization might
apply to them generally. Darnell and Vanek (1976) are also forced to
apply the concept of power broadly and inconsistently.
In contrast, Dahlstrom (App.: 14), approaching the problem from a
predominantly linguistic perspective, has argued against the view that the
animate gender perspicuously indexes a culturally salient category. She
observes that there is no central feature that motivates inclusion in the ani-
mate gender, and that there is thus no single cultural category, such as
power, that corresponds to the grammatical category. She argues that
there are, however, semantic links a m o n g animates, and that the set of
animates has some internal structure and cohesion.
Nevertheless a popular (or w e might better say vulgarized) Neo-
Whorfian view of Algonquian grammatical gender as giving a direct
insight into the Algonquian mental world persists, with the connivance of
some academics (App.: 15).

DESCRIPTIVE FACTS

Singular and plural

The Algonquian genders have different plural endings, but in most or all
nouns in most languages they do not have distinct singular endings:
 GRAMMATICAL GENDER IN ALGONQUIAN 199

(1) The marking of gender on nouns in four languages

INANIMATE ANIMATE
'song, songs' 'kettle, kettles'

Cree nikamon, nikamona askihk, askihkwak

Menominee neka-mwan, neka-mwanan ahkeh, ahke-hkok
Ojibwa nagamon, nagamonan akik, akikoog
Fox nakamoni, nakamonani ahkohkwa, ahkohkoki

Fox, which retains Proto-Algonquianfinalvowels in all nouns, has singu-

lar -i in all inanimates and singular -a in all animates.

Semantic fields
Leonard Bloomfield (1962:27-36) sorted all the animate nouns in
M e n o m i n e e into categories. H e distinguished between animate nouns
with "definable gender," those in "certain definable classes" that were
entirely animate, and animates of "special gender," which belonged to
"less definable or indefinable spheres of meaning," giving the following
lists:

(2) Categories of animate nouns in Menominee according to Bloomfield

(Definable gender)
Persons (including corpses).
Animals [including insects].
Spirits.
'Tree' and trees of full-sized species.
Trees of special shape, timbers, boards.
(Special gender)
Certain parts and growths of the human or animal body.
A few bodily secretions.
Animal hides.
The maize plant and its characteristic parts and products.
The wheat plant and its products,
Tobacco and terms applied to it; pipe, cigarette, cigar.
Raspberry and blackberry; their bushes and vines.
Various other plants and parts of plants.
Certain other natural objects
Shells, beads, and chinaware.
S o m e accessory garments and articles of personal adornment.
200 IVES GODDARD

Some other articles of intimate use.

Some other artifacts and products.
Moving machines, except for boats or ships.
The playing cards (except the ten-spot).
The categories in (2) give a general idea of the situation in all languages
though the specifics differ. Note, however, that the categories of "special
gender" include (or could include) just about everything except abstract
nouns and m a y b e structures: "certain" parts of the body, "a few" secre-
tions," "various" plants, "certain other natural objects," " s o m e " garments
and so forth, "some other" personal items, artifacts, and products.

Disagreements between languages

One measure of the partial arbitrariness of the animate category is the
extent to which languages differ on gender assignment. Note the follow-
ing examples:
(3) Nouns of different genders in different languages
'tobacco'
Animate: F asemawa, C cistemaw, M ne?nemaw, O asemaa
Inanimate: Mun kwsdhtew, Un kwsdtay, EAb watamaweyi
'com'
Animate: F atamina, M wapemen, O mandaamin
Inanimate: Munxwdskwim, Unxdskwim, EAbskdmon

2. Abbreviations used in this paper for language names before cited forms and seg-
ments: C = Cree; C h = Cheyenne; E A b = Eastern Abenaki; F = Fox (Meskwaki); M -
Menominee; Mass = Massachusett; M u n = Munsee; N U n = Northern Unami; 0 =
Ojibwa; P A = Proto-Algonquian; Sh = Shawnee; U n = Unami.
Abbreviations for grammatical terms: anim., A N = animate; coll. = collective; dim., DIM
= diminutive; E M P H = emphatic; EXPR.NEG. = expressive negative; H R S Y = hearsay;
inan., INAN = inanimate; obv., O B V = obviative; pi., PL = plural; Q U O T = quotative; sg.,
SG = singular.
The punctuation "!?" indicates the expressive negative sentence intonation, characterized
by overall raised pitch register and a high-low-mid contour on the last three voiced vow-
els. The sign "f" marks conjectural phonemicizations of words attested only in syllabic
writing. The sign " = " marks the attachment of clitics or independent words that are cliti-
cized.
In the interlinear glosses of relative roots, which bear a valence for an oblique comple-
ment, curly braces ({ }) enclose the descriptor of the type of obi iquc complement.
 GRAMMATICAL GENDER IN ALGONQUIAN 201

'bread'
Animate: Mpahkisekan 'wheat, flour, bread', O bakwezhigan 'bread,
flour'
Inanimate: Fpahkwesikani 'flour', wepiskwe-htehi 'bread'
'tree'
Animate: C mistik, M me?tek, O mitig, U n hitukw, M u n mihtdkw
Inanimate: F mehtekwi, Sh m?tekwi, Mass mahtakw ((mehtukq), etc.)
'heart'
Animate: U n wte- (wte-ha 'his heart')
Inanimate: Fmetehi, C miteh, Mmete-h, O ode' 'hisheart'
'flower'
Animate: Munpepdxkwales, Unb-tae-s
Inanimate: F peskonewihi, M wa-?sa-hkonawst, O waabigwan, E A b
pskwdhsawe
'paddle'
Animate: C a/?oy
Inanimate: F apwihi, Mpih, O abwi
'spoon'
Animate: Cemihkwa-nis,Une-mhb-nas,Mass&a«am(zaA:pl., EAbewiA:-
wa«. AThebiyo-
Inanimate: F e-mehkwa-hi, M e-meskwan, O emikwaan, C h hdmesko
'nail (for nailing things)'
Animate: M u n msko-s, U n mako-s
Inanimate: Fsakahkohikani, Csakahikan, Msakahekan, Ozaga'zga«

The animacy of tobacco figures prominently in the literature (Bloomfield

1962:32; Darnell and V a n e k 1976:169; Straus and Brightman 1982:105,
130; Dahlstrom 1995:56), yet in Eastern Algonquian tobacco is inanimate
(3). T h e cultural role and status of tobacco differed little a m o n g the
Algonquian peoples, and this divergence in gender shows that the linguis-
tic category of animate does not transparently label a corresponding
aspect of Algonquian world-view. T h e list in (3) could be greatly ex-
tended.

3. Written (kenamoog), (kunnomaog). Mass l&nama- is as if < P A **keOamaw-a

'thing constrained by biting', a derivative from the passive inflection of a transitive ani-
mate verb (cf. Goddard 1990:473, ex. 114), showing that when it was coined the existing
word for 'spoon' must have been animate.
4. This is the usual form in Fox texts; amehkwahi, much less c o m m o n in texts, seems
to be the universal form today. The non-diminutive emehkwani, found a few times in
texts, is an archaism.
202 IVES GODDARD

Inanimates used as animates in Menominee and Cree

Another area of interest is the occasional shift of inanimates to animates
in traditional stories, a p h e n o m e n o n which has played a leading role in
attempts to explain unexpected animates:
Any object which partakes, for the nonce, of personal qualities, especially
in the way of understanding speech and speaking, is in that context animate;
so, for instance, the normally inanimate nouns [M]penehkwan 'comb' in
the story of the Magic Flight and [M] we-nekan 'skull' in die Rolling Skull
story. (Bloomfield 1962:28.)
Hockett (App.: 10), as mentioned, also makes a point of this, and Straus
and Brightman (1982) provide extensive discussion of relevant texts, two
of which I have reprised.
In the more transparent cases, as in the M e n o m i n e e excerpts in (4)
and (5), an inanimate noun simply shifts to animate gender at some point
in the narrative. In (4) the skull shifts from inanimate to animate when it
becomes a Rolling Skull, a conventional ogre in myths. It is referred to as
inanimate in the quoted speech (as shown by the inflection of the verb for
an inanimate object and by the inanimate demonstrative pronoun), and in
thefirstclause of the following sentence (indicated by the verb ending). It
then shifts to animate in the main clause of the same sentence, as shown
by both the verb ending and the demonstrative:

(4) Menominee, "The Rolling Skull" (maskwawanahkwatok in Bloomfield

1928:430-1; successive sentences)
.1 "anowkch kekenuanenemuaw
in.vain of.course I.forbade.you.(pi.)
pahpenotamek eneh we-nekan."
(that).you.PL.toy.with.lNAN that.INAN skull
.2 ke?ceh pes-awek,
near when.lNAN.was.hither. {so}
enewen-peh pes-nekamit en oh we-nekan: ...
then.QUOT AN.SG.came.singing that.AN skull
.1 '"In vain I forbade you to abuse that skull [INAN]!'

5. The transcription of (4) and (5) has been phonemicizcd following Bloomfield (1962,
1975), but nor normalized.
 GRAMMATICAL GENDER IN ALGONQUIAN 203

.2 W h e n it [INAN] had come near, then that Skull-Being [AN] sang as it

came: ..."

In (5), the c o m b shifts from inanimate to animate w h e n it becomes a

talking c o m b (cf. Straus and Brightman 1982:118). This happens in the
course of a w o m a n ' s instructions to her c o m b after she has shut the Roll-
ing Skull in a sweatlodge. In (5.1) the verb for 'she said to it' construes
the object as inanimate and an inanimate demonstrative is used. In (5.2),
after the quote in which the w o m a n instructs the c o m b to speak, 'she said
to it' is rendered with an animate object, and the demonstrative and the
noun have distinctively animate forms. Specifically, the object (5.2) is
marked as being obviative, a secondary third person category distin-
guished only for animates in Menominee.

(5) Menominee, same text (Bloomfield 1928:434-5; successive sentences)

.1 "sa?sh, 'nekata-kece^tam;pa-hkenenaP'!'
soon I.want.to.come.out open.the.lodge.for.me
kenaw-ekuah," eta-men
AN.SG.will.say.to.you.AN she.said.to.INAN
eneh ope-nehkwan:
that. INAN her, comb .INAN
.2 "penah nap, 'kani, nawe-naw;
do in.return wait after.a.while
neme-k-sa-we-ha-n,' kenaw-ena-w,"
I'm.combing.my.hair you.AN.will.say.to.AN.SG
ens-wen 'noh ops-nehkwanan.
she.said.to.AN.QBV this.AN.OBV her.comb.AN.QBV
.1 " 'Soon, "I want to come out now; open it for me!" it will say to you,'
she told her comb [INAN];
.2 'but do you then, "Just a moment, I a m combing m y hair," do you say to
it,' she told that C o m b [AN] of hers."
In more complex cases, as in the Cree text about a Rolling Skull in 6
(cf. Straus and Brightman 1982:115-7), there is not an abrupt shift but
rather fluctuation between the genders, and even violation of the usual
patterns of grammatical agreement. This excerpt begins with a reference
to the w o m a n restated as a reference to the head, which is atfirstinani-
mate (in (6.1)), judging by the demonstrative and the verb agreement. In
 204 IVES G O D D A R D

(6.2), although it has the same inanimate demonstrative, it shows agree-

ment with the verb as an animate. It reverts to a consistent inanimate in
(6.3), even with the verb 'to speak', and then is fully animate in (6.4).

(6) Plains Cree, Rolling Skull episode (Louis Moosomin in Bloomfield

1930:8-9, 15-16; abridged)
.1 ekwa awa iskwew, kitahtawe-
and this-AN woman, at.some.point
tohkapimakan oma pisisik mistikwanis.
IN A N . SG. opened .its .eyes this.INAN only little.head

.2 ekwa kitahtawe- ka-pikiskwet om ostikwan....

and at.some.point AN.SG.spoke this.INAN head

.3 wahyaw oci wapahtam oma pisisik mistikwan

far from AN.SG.saw.lNAN this.lNAN only head
eh-pikiskwemakaniyik. ...
iNAN.QBV.spoke

.4 " 'namew' kike- 'siyihkason, " itik

sturgeon you.AN.will.be.called AN.QBV.said.to.AN.SG
awa ostikwan awa ka-tihtipipayit.
this.AN head this.AN AN^SG.that.rolls
. 1 "And that woman [AN], that severed head [INAN] presently opened its
eyes.

.2 Then presently that head [INAN] spoke [AN]. ...

.3 From afar he saw that severed speaking head [INAN], ...

.4 "Sturgeon" will be your name!' he [the Great Serpent, obviative] told
that Rolling Head [AN].

Unshifted inanimates in Fox

Things are somewhat clearer in Fox, thanks to the overt gender-marking

in all forms (l).6 In the parallel episodes to those in the M e n o m i n e e and

6. The Fox texts cited were written by native writers; they have been edited from the
original manuscripts in the National Anthropological Archives, Department of Anthropol-
ogy, National M u s e u m of Natural History, Smithsonian Institution. The translations eiven
in double quotes are from ones by native-speaking bilinguals in the same collection iden-
tified by the name of the translator. ' ae
 GRAMMATICAL GENDER IN ALGONQUIAN 205

Cree stories (4-6), the words for 'head' and 'comb' do not shift gender.
The word for 'head' m a y , however, alternate with the n a m e of the Rolling
Skull, w h i c h as a m y t h character is necessarily animate. In the first
excerpt given, this n a m e appears in (7.2). T h e next line (7.3) starts off
treating the creature as animate, as the subject of two verbs, but then it is
specified with the inanimate noun for 'head', overtly inflected as such. In
the last line (7.4) the skull is consistently inanimate.

(7) "Rolling Skull W o m a n " (sakihtanohkwe-ha; ms. pp. 17-19, abridged)

.1 eh=wepitanekomikateniki i-ni owi-wani
rNAN.QBV.began.to.act.up that.INAN his.wife.OBV
owisi. ...
her.head.lNAN
.2 "... e-h=ayi-hkwici-='niya ce-kesi-hita,"
as.AN.SG.is.tired=that.AN AN.who.is.small
eh=ici='pi-='na we-wi-se-ha. ...
AN.SG.said=HRSY=that.AN Rolling. Skull. AN
.3 eh=konakwataminici kepa-hkwapisocini,
AN.QBV.chewed.through.INAN iNAN.PL.that.bound.AN.SG.in
eh=pitikakoci owi-si. ...
AN.QBV.entered.where.AN.SG.was head. INAN
.4 eh=nanahkohaki='pi owi-si.
AN.SG.repeatedly.met.]NAN.with.blows=HRSY head.INAN
i-ni e-h=ayoci pehkwiki-hi.
that.INAN AN.SG.used.lNAN war.club.lNAN
.1 'His [slain] wife's head [INAN] began to act up. ...
.2 "... as that little one is tired," the Rolling Skull [AN] said. ...
.3 She [AN.0BV] chewed through the (vines) that he was held in by and
[AN.OBV] came in to where he [AN] was, the head [INAN] did. ...
.4 H e repeatedly beat back the head [INAN] with the ball-headed warclub.'
In the second Fox excerpt (8) the narrator seems to m a k e a joke out
of gender shifting, violating the expectations of the listener for comic
effect. A c o m b and what would conventionally be two acorns play the
role of the c o m b in the corresponding M e n o m i n e e episode (5):
 IVES GODDARD
206

(8) Rolling Skull episode in "The Apayashis" (sakihtanohhveha; ms. pp. 10-
13, abridged)
.1 i-ni = 'pi nahka penahakahi
then = HRSY also comb.DlM.lNAN
sakici eh-ahtoci, mehtekominehani-'nahi.
outside AN.SG.placed.lNAN. {somewhere} acoms.DlM.INAN =and
.2 "kih = tasi-nanahkomapwa," eh = itaki
you.AN.PL.must.answer.AN AN.SG.told.lNAN

i-ni penahakahi. ...

that.INAN comb.DlM.lNAN
.3 ini= 'pi, "pyeci-pahkenamawino.
then = HRSY come.and.open.it.for.me
.4 ini=yapi=mahaki e-h=ahtewaci asenyeki,"
now = PRESENT=these.AN AN.PL.have.gone.cold sweatlodge.stones.AN

eh^inaci.
AN..SC,.told.AN.QBV
.5 "ke-waki, nepenahahkwa=kohi," e-h = ikwici
wait I'm.combing.my.hair = certainly INAN.told.AN.SG

penahakahi,
comb.DlM.lNAN
.6 inini=nahka mehtekominani.
those.lNAN = also acorns.INAN8
.7 askacimekihi nahkaci='pi-='ni, '"sinakwa,
a.little.later again = HRSY = then What.the!
metopwawi- = nihka -kisi-penahahkwa-wane,"
not.yet = by.golly have.you.finished.combing.your.hair
eh=inaci.
AN.SG.told.AN.OBV
.8 " ke-waki = kohi, " eh = inici.
wait = certainly AN.OBV.said

7. 'Acorns (dim., inan.)'; could also be taken as 'acorn (dim., anim. obv.)'.
8. 'Acorns (inan.)'; could also be taken as 'acorn (anim. obv.)'.
 GRAMMATICAL GENDER IN ALGONQUIAN 207

.9 "pehki nina ma-hani ni-nesani sasakya-wani

really I these.lNAN my.hair.lNAN.PL iNAN.PL.are.tangled
ki-sakoci," eh=ici='pi=]na mehtekomine-ha. ...
extremely ARSG.said=HRSY=that. AN acom.DiM.AN

.10 "si-', 'ke-waki,' ketene=ma-h=e-yi-ki=ni-na,"

Gee wait I.told.you=you.see=as.well=I
e-h^ikwici^meko. ...
INAN.told.AN.SG = EMPH

.11 "'sina-', ke-waki=koh nepenaha-hkwa,"

what.fhe! still=certainly I'm.combing.my.hair
eh=iyo-mikateniki penahakani.
iNAN.OBV.said comb, I N A N

.1 Then she also placed a little comb [INAN] outside, and some little acorns
[INAN] (or a little acorn [AN]).

.2 "You guys must answer him," she told that little comb [INAN]. ...
.3-4 Then, " C o m e and open it for me. These sweatlodge-stones have gone
cold now," he (Rolling Skull [ A N ] ) said to her [AN.OBV].
.5-6 "Wait, I'm combing m y hair, I tell you," the little comb [INAN] and
those acorns [ I N A N ] (or that acorn [ A N . O B V ] ) told him.
.7 A little later then again, "Hey, for Pete's sake, haven't you finished
combing your hair yet?" he said to her [AN.OBV]. ...
.8 "Wait, I tell you," she [ A N . O B V ] said.

.9 " M y hair is really all badly tangled," said that little acorn [AN]. ...

.10 "Gee, /told you wait, too," it (or they) [INAN] said to him, insistently.

. 11 "Well, I'm still combing my hair, I tell you," the comb [INAN] said.

C o m b s and acorns are both grammatically inanimate in Fox, and in (8)

they appear to be, and to remain, inanimate until line (8.9), w h e n what the
listener would have thought were inanimate plural acorns suddenly sur-
faces unambiguously as an animate singular. T h e writer exhibits the same
sort of playfulness elsewhere as well, as with the appearance of the word
e-yi-ki 'too' in (8.10). T h e word-play in (8.9) is possible because of the
h o m o p h o n y of the endings for obviative singular and inanimate plural,
both being -ani, and of the corresponding demonstratives (both inini).
208 IVES GODDARD

A n y audience would take the 'acorn' words ending in -ani in lines (8.1) and
(8.6) as inanimate plurals, as they would know that the conventional talking-
acorn motif always involves a pair of acorns, although in the particular syn-
tactic contexts here it just happens to be the case that the two forms in -ani
would be equally grammatical as obviative singulars. In both sentences the
'acorn' word is conjoined with and follows the inanimate singular 'comb'
word, and since verbal agreement is always with the nearest of two conjoined
nouns of different genders (Dahlstrom 1995:62-64), or with thefirstif they
straddle the verb, the inflections used in (8.1) and (8.5-6) would be consistent
with a second conjoined noun of either gender. The animate gender is used for
just one line (8.9), after which inanimate agreement reappears, an indication
that the animate gender was, so to speak, not to be taken seriously. Note that
there is no motivation for the acorn to be animate in thefirstline; here the
word must be taken, onfirsthearing, as inanimate plural. Note also that
the c o m b is always inanimate both in form and in agreement, even when
speaking and spoken to. (The animate verbal inflection in (8.4, 8.7) and
(8.8) reflects the point of view of the Rolling Skull.)
In fact, in Fox inanimates seem to be freely assigned the powers of
speech, comprehension, and thought without shifting gender (9-13).
Verbs normally used only with animate subjects derive special forms for
use with inanimate subjects with the suffix -mikat (7.1, 8.11, 10.2, 11,12
[twice]); action by inanimate subjects on animate objects is indicated by
the use of the theme sign -ekwi (9.2, 13.1, 13.2).9
In (9) inanimate wind speaks:
(9) "The One That W a s Blessed by the White Buffalo" (Alfred Kiyana, ms., p.
160)
.1 "si-, wenah=mani nemisami,"
Gee, I.see.now = this.INAN (is) my.sacred.pack.INAN
eh=isiteheci.
AN.thought
.2 "ehehe," eh = ikwici notenwi.
Yes INAN.said.{so}.to.AN.SG wind.lNAN

9. The t of -mikat is replaced by h before k. The verb iN- (-eN-) 'say {so} to' has the
shape 0 (zero) before the inverse theme sign -ekw and its variants, and -ekw has its word-
initial variant ikw- after the modal prefixes (e-h= aorist, wih= future), which behave pho-
nologically like proclitic preverbs.
 GRAMMATICAL GENDER IN ALGONQUIAN 209

.1 "Gee, so this is m y sacred pack," he thought.

.2 "Yes," the wind [INAN] told him.
In (10) inanimate medicine is ascribed consciousness:
(10) Same text (pp. 245-6, abridged).
.1 na-hkaci=ki-nwa-wa mani kena-tawino-nwa-wi;...
also=you.PL this .INAN your.PL.medicine.lNAN
.2 kehke-netamomikatwi='yo=wi-na.
INANJ5G. is .conscious=for=but
.1 A n d this is your medicine [INAN]. ...
.2 For it [INAN] has consciousness.'
In (11-13) a sacred pack (misa-mi) is pleased, has the desire to lis-
ten, and speaks. In (11), mani 'this (inan.)' refers anaphorically to an
occurrence ofmi-sami in the immediately preceding context:
(11) " H o w the Fox Clan W a s Blessed" (Alfred Kiyana, ms. p. 160)
o-ni mani ahpeneci owiye-ha pinihtote
and.then this.lNAN always anyone.AN if.AN.SG.cleans.INAN
askote-wi, i-ni e-si-menwe-netamo-mikahki.
fire.INAN that.INAN (is) how.lNAN.is.pleased
"And if some one cleans up thefireplaceoften, this pack [INAN] will be
pleased very m u c h over it." (Translation by Thomas Brown.)
(12) Same text (p. 207).
ki-si-kehke-netamani manihi: nemisami
what.I.have.learned (is) this.lNAN my.sacred.pack.lNAN
e-h=aka-wa-tamo-mikahki wi-h =pesepesese-mikahki.
that. I N A N . desires that.rNAN.listen.to.people.repeatedly
"And I have found that m y sacred pack [INAN] desires to be listening to
people." (Translation by Thomas Brown.)
(13) Same text (p. 711)
.1 "... i-ni e-simekowaci='yo-we
thatlNAN (is) what.AN.OBV.told.AN.PL=PAST
na-hinahi a-cimohekowa-ci,"
time when.AN.OBV.instructed.AN.PL
210 IVES GODDARD

eh = ikwici omisami,
iNAN.said.to.AN.SG his.sacred.pack.INAN

.2 eh=cipinawemekwici.
and.lNAN.startled.AN.SG.by.what.it.said
.1 "... That's what he told them at the time when he instructed them," his
sacred pack [INAN] told him,
.2 and it [INAN] startled him.
In fact, sacred packs are always inanimate in Fox, despite being ascribed
all manner of conscious thought and action.
F r o m this Fox evidence it does not appear that gender shifts could
have played a large role in the constitution of the animate gender class in
Proto-Algonquian, where all nouns would have had the same overt gen-
der-marking endings as Fox.

SOME NEW PERSPECTIVES

In fact, in addition to the resistance to gender-switching in Fox narratives,
there are some other facts about the relationship between the two genders
that have not previously been mentioned, or not given full weight, in the
literature on the subject.

Gender shifts from animate to inanimate

The claim that an animate cannot shift gender to inanimate must be re-
evaluated. Clear cases of this shift can be found (14-21). For example, it
seems evident that the animate form in (14), which has cognates in most
Algonquian languages, represents the older usage:

(14) Fox maneto-wa AN 'spirit, god' -» manetowi INAN 'spiritual or sacred

power'10

In several other cases (15-17) the things designated by the inanimate

forms are demonstrably later than those designated by the animate forms:

10. Also cited by Straus and Brightman (1982:134) with no source given. Jones's
(1911:850) manetowi is a prenoun, with derivational suffix -i, and Bloomficld's
manetowi (Goddard 1994:90) was made on the basis of the possessed form nemanetomi
"my mystic power" (Michelson l925a:84.8-9), but the unpossessed inanimate noun is
attested several times in texts.
 GRAMMATICAL GENDER IN ALGONQUIAN 211

(15) Fox amo-wa A N 'bee, member of any species of the non-formicid aculeate
hymenoptera' —» a-mo-wi INAN 'honey'
Usable honey did not exist in North America until the arrival of European
honey-bees, but F amowa 'bee' has widespread cognates and is clearly
an old word. Therefore, 'honey' must have been named after 'bee' rather
than the other w a y around.

(16) O ishkodekaan AN 'fire-steel' —» O ishkodekaan INAN 'lighter'

The animate noun in (16) n a m e s the older fire-making device (Baraga
1850; quoted above), while the inanimate noun n a m e s the later one
(Nichols a n d N y h o l m 1995:69).

(17) M u n ntdUmo-ns AN 'my pet, dog, horse' —> M u n ntdlSmons INAN 'my
shaving horse'
The animate word in (17) is a possessed diminutive of Proto-Delaware
*albm 'dog' (< P A *aOemwa; cf. N U n (allum)). T h e inanimate word
refers to a type of low workbench of European origin, somewhat like a
cobbler's bench, that holds or supports a piece of w o o d while it is being
shaved or smoothed by the person w h o sits astride it.
The historically secondary inanimate nouns in (14-17) are all lexical-
ized, but there are also cases in the F o x texts of apparent nonce shifting
(18-19, 20-21). These show that the gender shift from animate to inani-
mate continued to be a vital process in the language. The inanimate form
of 'grandfather' (18-19) is especially noteworthy and unarguable.
(18) F kemeso-mesena-na A N 'our grandfather' -> F kemesomesena-ni INAN
'our "grandfather" (applied to an inanimate object)'
This gender shift is found in the following textual passage (19); the refer-
ence is to a ceremonial pole, designated in the text by the inanimate noun
F mehtekwi 'tree, stick, pole':
(19) "The Eagle Clan Redstone Pipe Ceremony" (Alfred Kiyana, ms. p. 56)
neya-pi=meko e-nahkate-kehe wi-h = inahkate-wi
same.as.before=EMPH the.way.INAN.stood iNAN.SG.will.stand. {so}
mani kemeso-mesena-ni
this.lNAN our.(incl).grandfather.lNAN
"This grand-father [INAN] of ours will go back and stand just where it
stood." (Translation by Thomas Brown.)
 DD R

e ond a (20-2 1) in ol a gend r hift in a noun th t i oth n i

ry con istently animate in Fo and at lea t om oth r languag :
(20) F a·teso ·hka ·kana A sacred story; myth character' a·te o·hka·kani
r AN sacred story (about an inanimat object)'

The textual pas age that atte t this shift i in (21 ):

(21 ) "Sky Sacred Pack ' (Alfred Kiyana, ms . p. 1)
ki· vekwi-mi·sa ·mi a·teso·hka·kani manr.
sky- sacred .pack.INAN sacred.story.INAN this.INAN
mekesiwaki.
Eagle.Clan.members.AN
'This [INAN] is the sacred story [INAN] of the Sky Sacred Pack [INAN] ofth
Eagle Clan [AN].'
The compound noun ki·sekwi-mi·sa·mi 'Sky Sacred Pack' in (21) is yn-
tactically a nominal adjunct to the noun a·teso·hka·kani sacred story', in
the semantic role of possessor. Nouns are not inflected for inanimate po -
sessors in Fox. From its shape, ki·sekwi-mi·sa·mi could be a prenoun com-
pounded with a·teso·hka·kani, but this can probably be ruled out. For one
thing the sacred pack is what the story is about, and it would be incompatibl
with this topical prominence for the initial mention of it in the text to be as a
first-compound member. Secondly, clans are customarily spoken of a po -
sessing a sacred pack rather than simply its associated origin story, o
nz ekesiwaki 'Eagle Clan members' must in turn be the nominal-adjunct
possessor of the sacred pack, and for this to work syntactically ki- "'ekwi-
mi· "'a·mi mu t be a noun. In either case (though more easily under the
analy is here proposed) the inanimate gender of a·te o·hka ·kani in (21)
seems to be motivated only by the gender of mi·sa·mi 'sacred pack'.

Inanimate collectives
If, instead of trying to imagine what the significance of the genders, espe-
cially the animate, migh t be in isolation from each other, we look for
cases of contrast in function, a new perspective is gained. For example,
there are a number of cases (e.g., 22-31) in which an inanimate noun is
used w ith a general or collective reference to things that are referred to
indi viduall y as animate, typically with the same noun ste m :
 GRAMMATICAL GENDER IN ALGONQUIAN 213

(22) F soniyahi INAN 'silver, money'; so-niyaha A N 'a coin, a bill', pi.
soniyahaki 'coins, bills, money'
(23) F owiyasi INAN 'meat, flesh'; owi-ya-sa AN 'a piece or cut of meat'
(24) F owi-nenwi INAN 'fat (generic)'; owi-nenwa AN 'a piece of fat; bear
(25) F anake-hkwi INAN 'bark'; anakehkwa AN 'a piece of bark used (or
intended for use) as lodge covering'
(26) Un kekw INAN 'wampum (coll.)'; kekw AN 'a wampum bead', pi. kekok
(27) Sh nileci I N A N 'my hand'; nileca A N 'myfinger'(Alford 1929, John
20:2s)11
There are even inanimate collectives that refer to animals (28-29) and
people (30-31):

(28) F micipehi INAN 'game (collective)'; cf. mi-cipe-ha AN 'a game anima
micipehaki
(29) F owiyehe-hi INAN 'animals, small game (coll.)'; cf. owiye-he-ha AN
(small) animal'
(30) F cinawetiweni INAN 'kin, relatives (coll.)', kecinawe-ti-weni 'your (sg.)
kin'
(31) Mass ndpamifdwak INAN ((nuppomeruonk)) 'my posterity, my
descendants'
The specifics of the use of gender and number with these nouns show a
fair amount of variety, even between nouns that are semantically close.
For example, the inanimate word for 'meat' (23) is m u c h more c o m m o n
than the inanimate for 'fat' (24), since certain kinds of fat, notably bear
fat, always seem to be animate.
Not all collectives are inanimate, however. C o m m o n animate collec-
tives in Fox are the following (32-34):
(32) F maskoci-sa AN 'beans (coll.), a bean', dim. maskocise-ha AN 'a sma
amount of beans; a single bean'
(33) F mesikwa A N 'dried c o m (coll.), uncooked or cooked; a kernel of dried
com', pi. mesi-kwaki 'dried com'; dim. mesikoha 'a small amount of dried
com (coll.); a kernel of dried com', dim. pi. mesikohaki 'a small amount of
dried corn'

11. The genders are reversed in Voegelin (1938-40:346); this is most likely simply an
error.
21 4 I G OOD RD

(34) F ata·mina A 'corn (generic), uncooked corn (coiL)', pl. atawzinaki ~om
(col i.) · dim. ata ·mine·ha ·a small amount of corn; a kernel of corn dilll.
pi. ata·mine· haki 'a small amount of corn; kernels of corn'
In these cases also each word has its own pattern of use. The word for .
dried corn ' (33), the usual word for corn as food, most frequently app a
in the plural, but the word for 'bean, beans' (32) never does, ev n wh n
individual beans are counted, and the generic word for corn (34) is often
singular.

Stems with gender contrasts

The collectives in (22-29) suggest approaching the question of the mean-
ing of the animate category by looking at further cases where the same
stem is used in both genders. Some examples of this, mostly from Fox,
are in (3 5 -41 ) :
(35) F aseni INAN 'stone'; asenya AN 'stone used in sweatlodge'
(36) F kehCipiso·ni, kehCipi·hi INAN 'belt'; kehCipiso·na, kehCipi·ha AN 'belt
(if a yam belt [44])'
(37) F apehkwe·simo·ni INAN 'head support'; apehkwe·simo·na AN 'pillow'
(38) F ahpisimo ·ni INAN 'thing for lying on'; ahpisimo·na AN 'mattress'
(39) F mehte/..:w i INAN 'stick, tree'; mehtekwa AN 'tree that is inviolate' 12
(40) F nehka·Ci JNAN 'my foot'; ohka·ta AN '(animal) foot (as food)' (also osita
AN, ositaya AN)

(41) M su·niyan INAN 'silver, coin, money, dollar'; su·niyan AN 'coin as

ornament'
In every case the nouns in (35-41) are inanimate when used for the
generic and the ordinary, and animate when used for the special or the
unusual. More information on the use of nouns that occur in both genders
in the same language would undoubtedly permit this list to be extended.
For example , M o·s ' canoe' is reported only to be inanimate, but in a

12. Attes ted as F mehtekwaki ' trees ', used for trees that are taboo as a source offirewood
in a te t dictated by Harry Lin coln (Michelson 1927:30.21 ); apparently equivalent to the
more common term mehtekwineniwa ' tree spirit'. The animate noun does not mean ' bow'
(Goddard 1994: l 0 I). The animate in thi s meanin g is a ghost created by a misi nterpretation
of nzehrelovan wi 'arrow' as spelled by Jones ( 1907:204. 12); sec th e correcti on in Goddard
( 1996a:6).
 GRAMMATICAL GENDER IN ALGONQUIAN 215

M e n o m i n e e text dictated b y Nehtsiwihtuk an elm-bark canoe is animate

when its manufacture is mentioned (ane-p-onan 'elm-bark canoe (obv.)')
but inanimate w h e n embarked in b y the ten youths (ota-nep-onemowa-w
13
'their elm-bark canoe') (Bloomfield 1928:256-7). W e have virtually n o
information o n such variation for any language.
T h e s a m e pattern is found in cases where semantically similar nouns
are of opposite gender (42-61):

(42) F mahkese-hi INAN 'moccasin'; a-kema AN 'snowshoe'

(43) FosehkwiweniINAN 'his spit'; otakikwaha AN 'phlegm', oskikoma AN
'snot'

(44) F asowanekwahatehi INAN 'crossbelt'; meskwa-swa-wa AN 'finger-

woven sash', "yam belt"

(45) F saponikani INAN 'needle'; samakwaya AN 'mat needle' (made of

flattened rib)

(46) F maskimotehi INAN 'bag'; pitanwana AN 'quiver', ahpwakanimoha AN

'tobacco pouch'
(47) Fci-ma-niINAN 'canoe';anake-weniINAN 'barkcanoe'; napehkwaniINAN
'sailboat, ship'; eskote-wi-hi INAN 'steamboat; railroad train';
kenwapihkatehiINAN 'freighttrain';
ata-pya-na AN 'wagon' (< '*sled, *drag'); so-skwiha AN 'sled, drag';
pinesi-pemipahoha AN, atamopina A N 'automobile'
(48) F a-cimo-ni INAN 'story, report'; atesohkakana AN 'sacred story; myth
character', "winter story"
(49) F mahkahko-hi INAN 'pail'; ci-keto-hi INAN 'tea-kettle';
ahkohkwa A N '(lidded) kettle; drum'; sesketoha A N 'cauldron'
(50) F matetehi INAN 'leggin'; asikasona AN 'sock, stocking' (< '*moccasin
liner')
(51) FmehtekwiINAN 'tree, stick, log';pasikahkwa AN 'board'

13. Bloomfield translates both occurrences as plural, but thefirstform is the object of a
transitive animate verb and the second form lacks any nominal-category suffix, either
obviative or inanimate plural. I am grateful to Marianne Milligan for bringing this passage
to my attention.
14. The older meaning is shown by the derivation from F atapyewa 'he drags (a sled,
drag)'; cf. O odaabaan A N 'sled, wagon, car'.
216 IVES GODDARD

(52) F mesehi INAN 'piece offirewood';kotawana AN, petewa AN 'large fire-

log'
(53) F asiskiwi INAN 'mud, clay'; wapiwena AN 'white clay', anemona AN 'red
ocher'
(54) F nepwami INAN 'my thigh, back of upper leg'; nepaya AN 'my femur,
front of upper leg'
(55) F nenepakayi INAN 'my forearm'; nehciwa AN 'my upper arm, biceps'
(56) F meskwi INAN 'blood'; atowa AN 'blood clot, coagulated blood'
(57) F nepi INAN 'water'; atekowa, atakowa AN 'wave'
(58) F mehkwami INAN 'ice surface'; mesihkwa AN 'ice, piece of ice
(59) F nekahkwani INAN 'my leg'; nenana AN 'my calf
(60) F nenekwikani INAN 'my arm' (also 'wing'); nehciwa AN 'my upper arm,
biceps'
(61) U n wikuwam INAN 'house';pimuwdkan AN 'sweatlodge'
A s the examples (35-61) show, the animate is consistently the spe-
cial or particular counterpart of the more ordinary, general, or inclusive
inanimate, or the animate is a part and the inanimate the whole. (Straus
and Brightman 1982:134 noted some of these patterns.) Animate are
fancy, special, or otherwise non-ordinary stones (35), belts (36), feet (40),
coins (41), footwear (42), needles (45), bags (46), logs and boards (51),
stories (48), structures (61), and so forth, in contrast to their plain or
generic counterparts; types of earth used for paints as opposed to ordinary
earth (53); things m a d e as pillows or mattresses as opposed to things
rigged up to be used for the same purposes (37-38); containers for cook-
ing as opposed to just for water (49); calf as opposed to leg (59); upper
arm as opposed to arm and to forearm (55, 60).
This is not the sort of distinction that leads directly to an unambigu-
ous assignment of gender. There would certainly be a cultural component
in making sleds special and canoes ordinary (47). A n d there are historical
connections that make wagons and hence cars animate, like sleds, and
trains inanimate, like canoes (47), to take only one case that cannot easily
be resolved by an appeal to power or inherent motion, or the like.

5. Cf. M mchkuam A N 'ice, piece or block of ice', INAN 'expanse of ice' (Bloomfield
 GRAMMATICAL GENDER IN ALGONQUIAN 217

B o d y parts present challenges to any analysis (54, 55, 59, 60). O n e

factor in this must be the fact that they can often be arranged in semantic
sets in m o r e that one w a y , and thus are susceptible to various analogies
and patterns.
Similarly, there are words for crops, skins, and trees of both genders
in Fox. T h e repartition into genders is probably consistent with the gen-
eral pattern, though where the line is drawn is arbitrary. For example, all
words for corn and beans, including all varietal names, are animate (32-
34), but other crops (62) are inanimate:

(62) F wapikoni INAN 'a squash, pumpkin'; manomini INAN 'wild rice'
(> 'rice'); oci INAN 'oats'; wi-ti INAN 'wheat'
In the case of skins (63) the critical factor in assigning gender is size,
large skins being as a rule inanimate, and small skins animate:

(63) Skins in Fox

Inanimate:
asayi 'skin of a larger animal'
nenoswayi, kohpici-nenoswayi 'buffalo hide'; mahkwayi 'bearskin';
mese-we-wayi 'elk skin'
wapi-ka-ka-nwikasewayi 'white-grizzly hide'; wapi-namipesiwayi
'white-underwater-panther hide'; wapi-mesikenepikwayi
'white-underwater-bear hide'
Animate:
asaya 'skin (of a smaller animal); buckskin'; owiyehaya, owiyehehaya
'skin of a small animal'; kekawe-waya 'skin with the hair on'
ka-ka-nwikasewaya 'grizzly-bear skin'
pesekesiwaya 'deerskin'; ayapewaya 'skin of a buck'; ke-takenehaya
'fawnskin'
mahwewaya 'wolfskin'; esepanaya 'raccoonskin'; monanehaya
'woodchuckskin'; amehkwaya 'beaver skin'; meswehaya 'rabbitskin';
ketatewaya 'otterskin'; ocekaya 'fisher skin'; mikaya 'mink skin';
asaskwaya 'muskrat skin'; kwakwinohaya 'groundsquirrel skin';
manetowaya 'snakeskin'; kiyo-te-waya 'serpent skin'
penesiwaya 'raptor skin'; e-he-waya 'swanskin'; wi-teko-waya 'owl skin'
apenohaya, apenohehaya 'skin of a human baby'

16. These three skins, from ancient manitou creatures, are mentioned in only one myth.
17. Among the accoutrements of the myth character Turtle is a tobacco pouch made of
this (Jones 1907:318.18).
218 IVES G O D D A R D

W e might observe that large skins can be directly used whole as robes or
mats, while small skins are typically variously worked on in being made
into clothing, bags, quivers, and other special items.
It is not clear w h y kakanwikasewaya 'grizzly-bear skin' should be
animate, especially given that wapi-kakanwikasewayi 'white-grizzly
hide' is inanimate. There are only two textual occurrences of
kakanwikase-waya, and it is possible that in both of them the gender has
been assimilated to that of the animates with which the skins in each
instance are identified. In one text the hero isfirstdescribed as using a
grizzly bear he has slain bare-handed as a breechclout:

(64) "The Sky Sacred Pack of the Eagle Clan" (Alfred Kiyana, ms. p. 42)
ini= 'pi inini ki-si-nesaci,
then = HRSY that.AN.OBV after.AN.SG.killed.AN.OBV

eh =pesinaci, eh=otasiyanici,
AN.SG.skinned.AN.OBV AN.SG.wore.(it.as).breechclout18

eh = wiwapisoci inini.
AN.SG.wrapped.himself that.AN.OBV
'Then after killing it he skinned it and used it for a breechclout, wrapping it
around him.'

Later in the text, w h e n the breechclout is mentioned, it is referred to by

the animate form of the word for the skin:

(65) Same text (ms. p. 219)

kewaki=wina=meko eh=menwikinici
still = but = E M P H AN.OBV.is.in.good.shape
kakanwikasewayani eh=otasiyanici.
grizzly.skin.AN.OBV AN.wore.(it.as).breechclout
'But the grizzly-bear skin was in good condition yet, being worn as his
breechclout.'

In the second case, one after another of a band of fleeing m e n throws

d o w n his quiver in front of a pursuing ogress in order to turn it back into

18. Verbs of possession may take the possessed item as a syntactic (secondary) object but
are not inflected for this object, which m a y be animate or inanimate; thus in the interlinear
the gloss for the object is parenthesized and not specified for gender. The instrumental
inini at the end of the sentence is taken as a nominal adjunct, for which the verb bears no
valence.
 GRAMMATICAL GENDER IN ALGONQUIAN 219

the animal it w a s m a d e from. Usually the quiver is referred to as being

made of the animal:

(66) "Giants" (Charley H. Chuck, ms. p. 12)

e-h=pakina-ci na-hka kotaka mese-we-wani
AN.SG.threw.AN.OBV.down also other.AN elk.AN.OBV
we-pi-tanwa-nita.
AN.SG.who.had.(it.as).quiver
'And the next one, whose quiver was made of an elk, threw it down.'
In two cases, however, the quiver is referred to as being m a d e of the skin
of the animal (67, 68):

(67) Same text (ms. p. 15)

"pena=ni-hka-='na kepi-tanwa-na pakisi,"
please=by.golly=that.AN your.quiver.AN throw.AN.down
e-h=ineci maci-hkiwesa.
AN.was.told Elder.Brofher.AN
kohpici-nenoswayani= 'yo=ke-hi e-h=opi-tanwa-nici.
buffalo.hide.AN.OBV=for=moreover AN.SG.had.(it.as).quiver
'"Better throw down your quiver," Elder Brother was told. N o w , the quiver
he had was of buffalo hide.'
(68) Same text (ms. p. 15)
nya-nano-nameki19 pye-hpahota
fifth AN.SG.who.comes.mnning
e-h=pakina-ci opi-tanwa-nani,
AN.SG.throws.AN.OBV.down his.quiver.AN.OBV
ka-ka-nwikase-wayani.
grizzly .skin.AN.OBV
'The one runningfifthfrom the end threw down his quiver, (made of) a
grizzly-bear skin.'
20 •
Since the words for buffalo hide are otherwise always inanimate, it
seems clear that the nonce occurrence of the animate gender of kohpici-
nenoswayani 'buffalo skin (obv.)' in (67) is due to attraction, perhaps to

19. Chuck writes nyananonameki 'fifth' with the numeral 5.

220 IVES G O D D A R D

the gender of pitanwana 'quiver' as m u c h as to that of the animal it is

made from. A striking case of gender attraction of this kind is found in an
archaic song text (69). Here moccasins, which are otherwise always inan-
imate, are referred to as animate, the gender appropriate to the material
from which they were made:
(69) "The Eagle Clan Redstone Pipe Ceremony" (Alfred Kiyana, ms. p. 159)
kiyo-tewayaki nemahkesenaki.
serpent.skins.AN.PL my.moccasins.AN.PL
'Of serpent skins are m y moccasins.'
The archaism of the song line in (69) is clear from the occurrence only
here of a form of the word for 'moccasin' with a stem F mahkesen-,
directly continuing P A *maskesen-. This stem was otherwise replaced in
the noun paradigm by mahkeseh-, in origin a hypocoristic diminutive
(Michelson 1925b:377; Goddard 1993:218), and survives only in derived
forms: mahkesenehke-wa 'she makes moccasins', nenotewimahke-
senewa 'he wears Indian moccasins', etc. Given cases like (67) and (69),
then, it is reasonable to a s s u m e that the animate gender of
kaka-nwikasewayani 'grizzly-bear skin (obv.)' in (68) is also a case of
contextual attraction. It is not clear, however, h o w to account for which
way the attraction goes, and a full account will require further examples.
Trees are another difficult set:
(70) Trees in Fox:
Inanimate:
mehtekomisi 'oak'; pehki-mehtekomisi 'white oak'; mahkatewikanohi
'black oak'; mi Simisi 'burr oak'
sasapihkahimisi 'willow'; pehkiwehimisi 'cherry tree'; kisowahkwi
'sycamore'; ani-pi 'American elm';peskipe-hi 'hickory'; wikopimisi
'basswood'; asenamisi, asanamisi 'sugar maple'; atopi 'alder';
meskwa-pi-misi 'red willow'; pakanahkwi 'walnut tree'

20. In the texts edited (or partially edited) so far by Lucy Thomason (to w h o m I am grate-
ful for drawing together examples) and myself there are 37 occurrences of inanimate
nenoswayi 'buffalo hide, robe' (including plural, diminutive, and possessed forms), two of
the younger term kohpici-nenoswayi, and two of wapi-kohpici-nenoswayi 'White Buffalo
hide'.
 GRAMMATICAL GENDER IN ALGONQUIAN 221

Animate:
e·skipakipye·ha 'evergreen'; sekwa·hkwa 'white pine';
meskwa·wa ·hkwa 'cedar';papaka·takwa 'cedar bough'' 'evergreen tree'
(Siebert 1967:25)
asa·siko·ha 'slippery elm'; mi·twi-wa 'quaking aspen' (Siebert
1967:31 ), "cottonwood" (Goddard 1991: 170); wi-saka·hkwa 'black
ash'; mya·nano·ha 'hop hombeam'
These lists are not complete, and in addition Adeline Wanatee accepted
some ordinarily inanimate trees (as well as certain other words) as ani-
mate when used "in the winter story" or "addressed religiously," for
example sasapihka·ha 'willow' and those given by Dahlstrom (1995:58).
In most Algonquian languages all trees have become animate, but Fox,
Shawnee, and Massachusett preserve the older pattern, in which the
generic term was inanimate (3) and the species names are found in both
genders. Oaks are inanimate and evergreens animate, but beyond that
generalizations are unclear.

Inanimate as a grammatical category

In addition to the inflection of inanimate nouns and the agreement of
other words with them, Fox also uses the inanimate as the unmarked gen-
der in grammatical functions unassociated with any noun either expressed
or implied. In such cases the gender cannot be taken as in any way refer-
ring to or indexing a category of world view or of cognitive or cultural
processing.
The inanimate indefinite pronoun ke·ko·hi 'something, anything;
thing(s)' may be used where the reference can only be to unspecified ani-
mates (cf. Dahlstrom 1995 :64), as in (71 ), where a man admits to his wife
that he did not kill an enemy while in a war party.
(71) "Wapasaya's Younger Brother" (Alfred Kiyana, ms. p. 43)
a·kwi=ma·h=ni·na ke·ko·hi nehtoya·nini.
not =you.see =l something l.did.(not).kill.INAN
'I have not killed anything.' (Translation by Thomas Brown.)
The inanimate pronominal inflection may be used the same way, as in
(72), where the reference is to killing animals:
(72) "The Story of One Blessed by an Evil Spirit" (Alfred Kiyana; ms. p. 221)
222 IVES GODDARD

mese = meko netawenetaki=meko

any = EMPH INAN. SG.that.AN.SG. sought = EMPH
eh =anemi-=meko -nehtoci.
AN.SG.went.on = EMPH -killing.INAN
'He went on killing anything he wanted to.'
The fact that the reflexive pronoun (e.g., F owi-yawi 'his body, him-
self, him') is grammatically an inanimate singular mass noun (Dahlstrom
1988:182-5) entails that periphrastic reflexive sentences always involve a
transitive verb with an inanimate object that is referentially identical to
the animate subject:

(73) "Shooter, and his Grandmother and Grandfather" (Alfred Kiyana, ms. p.
135)
eh=nehtonici owi-yawi,
AN.OBV.killed.lNAN his.body.lNAN
eh=nesetisonici.
AN.OBV.killed.self
'He (obv.) killed himself, he (obv.) committed suicide.'
For a number of verbs this construction is the only one in which the tran-
sitive inanimate stem is attested.
Yet another indication of the abstractness of gender as a grammatical
category is the use of inanimate demonstratives with emphatic pronouns
in equational sentences, which state the equivalence of two noun phrases.
In Fox equational sentences have no copular verb or particle. The given
(presupposed) and n e w terms of the equation most often occur in the
order given-new (9.1, 11, 12, 13.1, 74, etc.), but they m a y also occur in
the order new-given (69), or either term m a y be discontinuous with the
other interposed (10.1, 21). W h e n the n e w term is not sentence-initial it
m a y be preceded by a supernumerary demonstrative pronoun, like the
first i-ni in (74):

(74) "The Owl Sacred Pack of the Fox" (Alfred Kiyana, ms. p. 99)
mani- = 'ni- = 'ni natawinoni.
mani ini ini natawinoni
this.lNAN that.lNAN that.lNAN medicine.lNAN
"This [that we now see] is that [previously discussed] medicine.'
 GRAMMATICAL GENDER IN ALGONQUIAN 223

Emphatic pronouns, the set with ni-na T (Dahlstrom 1988:166-177),

despite ending in -a, are not animate nouns; they behave morphologically
like invariant particles. A n d , in fact, in equational sentences, including
simple presentational statements, the demonstrative pronoun that is used
with emphatic pronouns is the inanimate singular i-ni.
(75) "Wapasaya" (Alfred Kiyana, ms. p. 92)
ki-na= 'h=we-na-= 'ni ke-htena
ki-na =ihi =we-na i-ni ke-htena
you.SG =not = rather that.lNAN truly
'wa-pasaya' e-nenekil?
wa-pasaya e-nenekil?
Wapasaya you.who.are.called.{so} E X P R . N E G
ehe-he, ni-na=kohi i-ni 'wa-pasaya'
yes I=certainly that.lNAN Wapasaya
e-siki manetona-ki.
I.who.am.called. {so} manitou.land+LOC
" 'Is it true that you are the one who is called Wapasaya?'22 'Yes, I am tha
one, they call Wapasaya in the realms of the Manitous.'" (Translation by
Thomas Brown.)
(76) "Masahkamikohkwewa" (Alfred Kiyana, ms. p. 361)
ni-na=kehi-= 'nise-kihena-ni ...
nina =ke-hi ini se-kihena-ni
I =moreover that.lNAN I.who.scared.you
ni-na=ca-hi- = 'ni-= 'ni.
ni-na =ca-hi i-ni i-ni
I =so that.lNAN that.lNAN
T was the one that scared you,... I was the one.' (Translation by Horace
Poweshiek.)
The use of inanimate i-ni with the emphatic pronouns can be explained
straightforwardly o n the basis of grammatical categories: the less marked
gender category is used in constructions involving particles that have n o
grammatical gender. This explanation also accounts for w h yfirstand sec-
ond person participles (like those in (75) and (76)) have the inanimate sin-

21. I owe examples (75) and (76) to Lucy Thomason.

22. Or maybe better: 'So you're really the one called Wapasaya, are you.'
224 1 ODDARD

gular uffi -i and not th uffi -a amm. g. and -iki anim. l. that ar
u ed to mark anim t participial h ad . ut th u of th inanim t
would mak no en if the g nders actually encod d culturally d fin
emantic cat gorie , or if the animat ere u d in ppo ition with a
local pronoun " as Hockett phrased it ( . pp.: 1 . Th p r nal pr n un
particularly those of the fir t and cond p r n ar th high t ntiti
on any hierarchy ofinher nt animacy ( ilv r tein 1976:122· nv nit ·
1971:195-204 217 -222; Hockett 1966:60) and the fact that in rtain
constructions they ar construed a inanimate can only m an that g nd r
is a grammatical category, govern d by grammatical rul r th r th n y
culturally defined semantic principles.

CO CLUSIO

The basic meaning of the animate gender is a function of the contra t with
the inanimate ge nd er. Looking at the animate by th m lv and
attempting to connect the dots doe not reveal it. But wh n large number
of eemingly arbitrary animates are examin d together with semantically
close inanimates, patterns of contrast are revealed that define the emantic
corr lates of the animate category. The constant feature eem to be clo
to what the s venteenth-century Jesu it linguists lab e led ' nobl a
opposed to ' base', though without a necessary component of e t m. W
might update these term.s as, simply, high and low. Th animat i th
high gender, and the inanimate is the low gender. The assignm nt f
humans, animals, and spirits to the high gender would obviously be aut -
matic, as would be the assignment to it of perhaps a £ w other cat gories
of things . But the application of the semantic opposition between the gen-
ders to many specific cases involv s a cultural component that can vary
and that leads to different gender-c lass membership in different spe eh
communities.
The Algonquian genders do not, however, reflect a "cognitive ut-
look [that] is a radical departure from the framework of our thinking,"
still less does the list of animate nouns give us a li st of "the other selves"
in the Algonquian environment (Hallowell, App.: 9). On this point the
missionaries were right. The speakers do not regard the unexpected ani-
mates as a class, as living or powerful; rather, "the Indians treat [them] in
 GRAMMATICAL GENDER IN ALGONQUIAN 225

their language like substantives that signify living beings" (Baraga

1850:18 [emphasis added]).
Native-speaker attitudes appear to be similar, to judge by conversa-
tions I held on this subject with Martha Ellis about Unami in the 1960s,
with Rebecca Snake about Munsee in 1981, and with Adeline Wanatee
about Meskwaki in the 1990s. Some things regarded as spirits or as hav-
ing spiritual power, either inherently or for the nonce, are treated as ani-
mates. But most unexpected animates, though treated in the language like
things that are alive, are not believed to be alive. In discussing the phe-
nomenon, the usual and convenient explanation given by speakers is
something like, "you talk about it like it's something living," or "it sounds
like it's alive," expressions which make it clear that such animates are not
regarded as living. For example, the appearance in conversations some
years apart of English bicycle used as an animate in a Munsee context and
of animate F otakikwa·ha 'phlegm' occasioned similar discussions, in
which the agreement requirements of these words were pointed out with
some amusement, but there were no claims of deeper significance.
Those who have thought that decontextualized collections of animate
nouns directly reveal the mental culture of the Algonquian Other on its
own terms seem to have been describing instead a projection of their own
culture. Uncritical cultural relativism, made more extreme by unexorcised
Neo-Whorfianism, has produced an interpretation of the Algonquian
world-view that is as insupportable on the facts as it should be inconsis-
tent with common sense, though it may seem to confirm and affirm the
exoticism of the Algonquian mind in a way that, however circularly, vali-
dates the approach taken.
Still, the Algonquian gender system does have something to tell us
about the cognitive structures employed by Algonquian speakers, but in a
subtler, less expected, and perhaps more interesting way than heretofore
assumed.

APPENDIX: DESCRIPTIONS OF ALGONQUIAN GENDER, 1634-2000.

(1) Paul LeJeune, 1634 (in Thwaites 1896-1901, 7:22-23 [modernized;
underlining added]):
"lis ont des verbes differents pour signifier I' action envers une chose
animee et envers une chose inanimee, encore bien qu'ils conjoignent avec
les chose animees quelques nombres des choses sans ame, comme le petun,
226 IVES GODDARD

les p o m m e s , etc."
(2) John Eliot (1666:9 [italics original]):
> , rx, f Animate.
"There be two forms or declensions of Nouns: | j n a m m a t e

"The Animate form or declension is, when the thing signified is a living
Creature.... Some few Exceptions I know."
(3) "Principes de la langue algonquine" (ca. 1662; Hanzeli 1969:103
[modernized; underlining added]):
"lis ont peu de noms. Les uns sont nobles, qui sont des chose vivantes ou en
estime,.... Les autres sont ignobles. qui sont des choses inanimees."
(4) Louis Andre, "Preceptes, phrases et mots de la langue algonquine
outaouoise pour un missionaire nouveau" (ca. 1688; Hanzeli 1969:122 and
facing photo [underlining added]):
"syluestres non habent genera diuersa sed nomina, pronomina, et uerba
nobilia et ignobilia."
(5) Louis Nicolas, "Grammaire algonquine" (ca. 1672-74; Daviault 1994:33
[underlining added]):
"Nom noble veut dire et se raporte aux mots qui signifient les choses
vivantes ou de consideration parmy les Sauvages. N o m ignoble: choses qui
n'ont point de vie ou de peu de consideration."
(6) Henry R. Schoolcraft (1852, 2:365-7):
[T]here is perhaps no feature which obtrudes itself so constantly to view, as
the principle which separates all words, of whatever denomination, into
animates and inanimates, as they are applied to objects in the animal,
vegetable, or mineral kingdom. ...
Nouns animate embrace the tribes of quadrupeds, birds, fishes, insects,
reptiles, crustacas [sic], the sun, and m o o n , and stars, thunder and lightning
[(]for these are personifiedQ],23 and whatever either possesses animal life,
or is endowed, by the peculiar opinions and superstitions of the Indians,
with it. In the vegetable kingdom, their number is comparatively limited...
It is at the option of the speaker to employ nouns either as animates or as
inanimates; but it is a choice never resorted to, except in conformity with
stated rules. These conventional exceptions are not numerous, and the more
prominent of them m a y be recited. The cause of these exceptions it is not

23. The original punctuation has been replaced, since the sense requires "for these are
personified" to be taken as a parenthetical phrase.
 GRAMMATICAL GENDER IN ALGONQUIAN 227

always easy to perceive. It may, however, generally be traced to a particular

respect paid to certain inanimate bodies, either from their real or fancied
properties, the uses to which they are applied, or the ceremonies to which
they are dedicated. A stone, which is the altar of sacrifice to their manitoes;
a b o w , so necessary in the chase; a feather, the honored sign of martial
prowess; a kettle, so valuable in the household; a pipe, by which friendships
are sealed and treaties ratified; a chum, used in their sacred and festive
dances; a medal, the mark of authority; vermilion, the appropriate paint of
the warrior; w a m p u m , by which messages are conveyed, and covenants
remembered. These are among the objects, in themselves inanimates, which
require the application of animate verbs, pronouns, and adjectives, and that
are thereby transferred to the animate class. ...
In a language which considers all nature as separated into two classes of
bodies, characterized by the presence or absence of life, neuter nouns will
scarcely be looked for, ...
Jean-Andre Cuoq (1891:88-89):
"Les Algonquins partagent les etres en deux grandes classes auxquelles on
est convenu de dormer le n o m de genre anime et de genre inanime."
William Jones (1911:761 [underlining added]; cf. 1904:385):
" A rigid classification of the objective world into things animate and things
inanimate underlies the whole structure of the language."
"... The distinction between the two opposing groups is not rigidly
maintained, for often an object regularly inanimate is personified as having
life, and so takes on an animate form. But permanent forms of lifeless
objects having an animate ending can not always be explained by
personification. The breaking-down of the contrast is best seen in the names
of plants; logically they fall into the inanimate class, but m a n y are used as
animate forms."
A. Irving Hallowell (1955:109):
"It has been said that the grammatical distinction between animate and
inanimate gender in Ojibwa speech is arbitrary and hard to master. It only
appears so to an outsider. Actually, it is precisely these distinctions which
give the Ojibwa individual the necessary linguistic cues to the various
classes of other selves that he must take account of in his behavioral
environment."
Hallowell (1976:362-3):
" W h e n evidence from beliefs, attitudes, conduct, and linguistic
characterization are all considered together the psychological basis for their
228 IVES GODDARD

unified cognitive outlook can be appreciated, even w h e n it is a radical

departure from the framework of our thinking. ... M o r e important than the
linguistic classification of objects is the kind of vital functions attributed to
them in die belief system and the conditions under which these functions
are observed or attested in experience. This accounts, I think, for the fact
that what w e view as material, inanimate objects - such as shells and stones
- are placed in an 'animate' category along widi 'persons' which have no
physical existence in our world view. The shells, for example, called
[miigis][,]24 on account of the manner in which they function in the
Midewiwin, could not be linguistically categorized as 'inanimate.' ... The
hypothesis which suggests itself to m e is that the allocation of stones to an
animate grammatical category is part of a culturally constituted cognitive
'set.'"
(10) C F . Hockett (1966:62):
"The genders of Algonquian are not balanced: the animate gender is, as it
were, absorptive. If an inanimate noun stands in apposition with a local
pronoun [=lst or 2nd person], or if that named by the noun takes on the
power of speaking or of being addressed (as in a story), or in any other way
is assigned powers usually associated with people, animals, and spirits, it
tends to become animate in that context and to be so treated syntactically
[=in agreement], sometimes even inflectionally. ... [T]here are routes for a
shift of gender from from inanimate to animate, but not the opposite. W e
m a y suspect that this ... accounts for the scattering of semantically
arbiUary and surprising animate nouns in the present-day languages."
(11) Regna Darnell and Anthony L.Vanek (1976:163, 177-8):
"Power involves ability and freedom to act and interact.... In the native
conception, it is power that moves the universe and that maintains its
balance; balance in this instance is not passive, but a dynamic tension of
forces. Such a feature will not account for the gender of every Cree noun; it
will, however, postulate a set of features which designate categories of
animate objects, and thereby delineate the structure of the semantic domain.
In any one case, the explanation m a y be somewhat arbitrary. But the weight
of a semantic domain emerges as a class, not as an item."
"... gender assignment in Cree ... deals with power to maintain and
balance the universe and to interact with persons and other interactive
beings. W h a t is involved goes to the heart of the division between the
realms of the natural and the supernatural. A n d this, in turn, is clearly a
cultural rather than a linguistic matter."

24. After the retranscribed word I have added a comma, since the phrase "on account of
the manner in which they function in the Midewiwin" modifies what follows.
 GRAMMATICAL GENDER IN ALGONQUIAN 229

"... the semantic domain does have a certain coherence, w h e n a culturally-

appropriate semantic feature is taken as the criterion."
(12) M a r y B. Black-Rogers (1982:67):
"To explain how it happened that a good number of unexpected and
assorted nouns came to have a stable A[nimate]-gender assignment, the
'power' hypothesis implies that at some time in the past some otherwise
inanimate and/or powerless entities were imbued with special qualities,
uses, or abilities that associated them with power. .. .this gender assignment
could have occurred initially through the phenomenon of gender shift."
(13) Anne Terry Straus and Robert Brightman (1982:99, 134):
"We hold that gender is overwhelmingly definable."
"The fundamental semantic contrast which corresponds to gender in
Northern Cheyenne and probably other Algonquian languages [i]s that
between powerfulness and powerlessness as attributes of referents."
(14) Amy Dahlstrom (1995:64-65):
"The animate gender of Algonquian languages is marked and the inanimate
gender is unmarked; furthermore, the animate category has internal, radial
structure, with the feature of [+animate] picking out the central members of
the category. The feature of power is not the central feature, but is an
important extension from animacy, motivating m a n y instances of animate
gender assignment. ... [T]here are other possible motivations for a noun
being assigned animate gender... [W]e cannot use gender morphology to
simply read off the power associations of objects in Algonquian culture."
(15) Alexander Stille (2000):
"For, example, M r . [Akira] Y a m a m o t o points out, in the Algonquin family
of languages, noun endings are divided into two basic categories: animate
and inanimate. So, while Romance languages separate nouns by gender, the
Algonquin sees the world in terms of things that have spirit and things that
do not. And, M r . Y a m a m o t o adds, "This is reflected in their culture.'"

REFERENCES
Alford, Thomas W., tr. 1929. The gospel of our Lord and Saviour Jesus Christ according
to Matthew translated into the Shawnee Indian language. Xenia, Ohio: Dr. W.A.
Galloway. [Cover title: The four gospels of Our Lord Jesus Christ in Shawnee Indian
language.]
Baraga, Friderik. 1850. A theoretical and practical grammar of the Otchipwe language.
Detroit: Jabez Fox.
230 IVES GODDARD

Bcnveniste, Emile. 1971. Problems in general linguistics, tr. by Mary Elizabeth Meek.
Coral Gables: University of Miami Press.
Black-Rogers, Mary B. 1982. Algonquian gender revisited: Animate nouns and Ojibwa
'power' - an impasse? Papers in Linguistics 15(1 ):59-76.
Bloomfield, Leonard. 1928. Menomini texts. Publications of the American Ethnological
Society 12.
— . 1930. Sacred stories of the Sweet Grass Cree. National M u s e u m of Canada Bulletin
60. Ottawa.
— . 1962. The Menomini language. N e w Haven: Yale University Press.
Cuoq, Jean-Andre. 1892-93. Grammaire de la langue algonquine. Transactions of the
Royal Society of Canada for 1891, vol. 9 (sect. 1):85-114 (1892), ...for 1892, vol.
10 (sect. 1):41-119 (1893).
Dahlstrom, A m y . 1988. Independent pronouns in Fox. In honor of Mary Haas: From the
Haas Festival Conference on Native American linguistics, ed. by William Shipley,
pp. 165-194. Berlin: Mouton de Gruyter.
— . 1995. Motivation vs. predictability in Algonquian gender. Papers of the 26th Algon-
quian Conference, ed. by David H. Pentland, pp. 52-66. Winnipeg: University of
Manitoba.
Darnell, Regna, and Anthony L. Vanek. 1976. The semantic basis of the animate/inani-
mate distinction in Cree. Papers in Linguistics 9(3/4): 159-180.
Daviault, Diane. 1994. L'algonquin au XVIle siecle. Une edition critique, analysee et
commentee de la grammaire du Pere Louis Nicolas. Sainte-Foy, Quebec: Presses dc
l'Universite du Quebec.
Eliot, John. 1666. The Indian grammar begun. Cambridge: Marmadukc Johnson.
Reprinted (with original pagination noted) as A grammar of the Massachusetts Indian
language, ed. and with notes and observations by Peter S. Duponceau and an intro-
duction and supplementary observations by John Pickering. Boston: Phelps and Fam-
ham, 1822.
Goddard, Ives. 1991. Observations regarding Fox (Mesquakie) phonology. Papers of the
22nd Algonquian Conference, ed. by William Cowan, pp. 157-181. Ottawa: Carleton
University.
— . 1994. Leonard Bloomfield's Fox lexicon: Critical edition. Algonquian and Iroquoian
Linguistics Memoir 12. Winnipeg.
— . 1996a. Leonard Bloomfield's Fox lexicon: Critical edition (1994)- addenda and corri-
genda as of April 1, 1996. Algonquian and Iroquoian Linguistics 21(l):6-7'.
— . 1996b. The description of the Native languages of North America before Boas. Hand-
book of North American Indians, v. 17: Languages, ed. by Ives Goddard, pp. 16-42.
Washington: Smithsonian Institution.
Hallowell, A. Irving. 1955. The self and its behavioral environment. Culture and experi-
ence, pp. 75-110. Philadelphia: University of Pennsylvania Press.
— 1976. Ojibwa ontology, behavior, and world view. Contributions to anthropology:
Selected papers of A. Irving Hallowell, pp. 357-390. Chicago: University of Chicago
Press.
Hanzeli, Victor E. 1969. Missionary linguistics in New France. The Hague: Mouton.
Hockett, C F . 1966. What Algonquian is really like. International Journal of American
Linguistics 32:59-73.
 GRAMMATICAL GENDER IN ALGONQUIAN 231

Jones, William. 1904. S o m e principles of Algonquian word-formation. American Anthro-

pologist, n.s., 6:369-411.
— . 1907. Fox texts. American Ethnological Society Publications 1.
— . 1911. Algonquian (Fox), ed. by Truman Michelson. Handbook of American Indian
languages, ed. by Franz Boas, pp. 735-873. Bureau of American Ethnology Bulletin
40, Part 1.
Michelson, Truman. 1925a. The mythical origin of the White Buffalo Dance of the Fox
Indians, together with texts on four minor sacred packs appertaining to this cere-
mony. Bureau ofAmerican Ethnology Annual Report 40:23-289.
— . 1925b. Notes on Fox mortuary customs and beliefs. Bureau of American Ethnology
Annual Report 40:351-496.
— . 1927. Contributions to Fox ethnology. Bureau of American Ethnology Bulletin 85.
Nichols, John D., and Earl Nyholm. 1995. A concise dictionary of Minnesota Ojibwe.
Minneapolis: University of Minnesota Press.
Powicke, Frederick James. 1931. Some unpublished correspondence of the Reverend
Richard Baxter and the Reverend John Eliot, the Apostle of the American Indians,
1656-1682. Manchester: Manchester University Press.
Schoolcraft, Henry Rowe. 1852. A n essay on the grammatical structure of the Algonquin
language. Historical and statistical information, respecting the history, condition a
prospects of the Indian tribes of the United States, ed. by Henry R. Schoolcraft, vol.
2, pp. 351-442. Philadelphia: Lippincott, Grambo.
Siebert, Frank T., Jr. 1967. The original h o m e of the Proto-Algonquian people. Contribu-
tions to Anthropology: Linguistics I, ed. by A.D. DeBlois, pp. 13-47. National
M u s e u m of Canada Bulletin 214.
Silverstein, Michael. 1976. Hierarchy of features and ergativity. Grammatical categories
in Australian languages, ed. by R.M.W. Dixon, pp. 112-171. Canberra: Australian
Institute of Aboriginal Studies.
Stille, Alexander. 2000. Speak cultural memory: A dead-language debate. New York Times,
30 September 2000, A17.
Straus, Anne Terry, and Robert Brightman. 1982. The implacable raspberry. Papers in
Linguistics 15(1):97-137.
Thwaites, Reuben G., ed. 1896-1901. The Jesuit Relations and allied documents. 73 vols.
Cleveland: Burrows Brothers.
Voegelin, Carl F. 1938-40. Shawnee stems and the Jacob P. Dunn Miami dictionary. Indi-
ana Historical Society Prehistory Research Series 1:63-108, 135-167, 289-341, 345-
389, 409-478. Indianapolis: Indiana Historical Society.