Semiotic Review 2: Monsters - article published March 2015

The Return of the Windigo, Again

Robert Brightman


Keywords: windigo; Peirce; metamorphoses; animism


I must thank Paul Manning, no enemy of irony, for remediating an unreconstructed structural reading of the Algonquian windigo in the other-than-structural (?) precincts of Semiotic Review [Brightman,“And Some Guys Dream Bad Things”]. Some people might take this Finnegan’s Wake-like resuscitation as occasion for reflecting on what specifically Peirce-inspired semiotic readings of windigo might consist of. A survivor of psychological anthropology (“culture bound syndromes”), functionalism, structuralism, cultural materialism, and even political economy, the windigo has yet to experience semiotic. I will briefly introduce the two here, gesturing wildly in brief transit at a few among many possible sites of connection. The piece presupposes some familiarity with Peirce, and those wanting initiation or tuning-up are referred to Parmentier (1994,) Lee (1997), and Short (2007).

Is Windigo Real?

Readers of Semiotic Review are no doubt asking themselves how the modernist realism of Peircian semiotic and pragmat(ic)ism comports with signs whose objects are conventions--conventions, moreover, which Peirce, had he been versed in windigos, would have dismissed as contracting fanciful relations with the realities.

Peirce offered varying perspectives on encompassment of conventions and non-existing entities by the “Real.” His conclusion that non-real or non-true objects determine signs just as their real and true counterparts do occurs in different phrasings (Parmentier 1994:19-22). Thus in 1903, lies, false propositions (“Napoleon was a lethargic creature“), literary fictions, and military commands (Peirce’s beloved and ubiquitous “Ground arms!”) posed no exception to the postulate that objects determine their signs (CP 8.178). The same year, Peirce took explication of rhematic symbols as occasion to affirm that existing knowledge of a non-existing entity makes a perfectly acceptable object. Common nouns stand for “general concepts” and their replicas stand for the latter’s instances. Replicas of “camel” index the general concept of camel only by “being really affected” by interlocutors’ shared knowledge of camels. This knowledge, is, in its turn, affected by “the real camel.” So also with “phoenix” where, however, conventional knowledge of phoenixes stands in, qua object, for the real entity itself: “For although no phoenix really exists, real descriptions of the phoenix are well-known to the speaker and his auditor, and thus the word is really affected by the Object denoted” (CP 2.178, emphasis added). Peirce (MS 634[1909]) later revisited the same argument with “witch,” whose “real [but not “altogether veracious”] object” is characterized--with particular relevance to windigo--by analogy to dream imagery whose contents “are not an appearance of objects that are Real” but which are real qua subjective experience (CP MS 634:26-27, cf. Peirce on “Reflections on Real and Unreal Objects” [CP MS 966], cf. Parmentier 1994:21).

Semiosis of Windigo Metamorphosis

Primary among Algonquian windigo schemata is the principle that people can turn into them. Let us look at this doctrine before contemplating its effects. At the heart of things is the (here in Woods Cree) ithiniwak ati-wiihtikoowiwak aaskaw ‘people sometimes become windigos.’ This is the proposition of which such particular dicent symbols as na-ati-wiihtikoowin ‘I will become windigo’ are referring replicas.

A stereotyped set of distal causes transform human individuals into windigos, and a stereotyped set of ideational-physiological symptoms are signs of these causes and of the concomitant progress of transformation. Leaving convention aside, for the nonce, distal causes, proximal symptoms, and metamorphosis events are indexical icons, contracting reversible sign-object relations with one another.

The four, sometimes blended, causes of metamorphosis were dream events, sorcery, spirit “possession,” and prior logistical (e.g., “lifeboat”) cannibalism. Of the symptom-indexes, the most windigo-specific are anthropohagous ideation or desire, freezing of the heart, and perceptions of humans as game animals. Also commonly described are anorexia (or its inverse, gluttony) and demeanors characterized as “melancholia” and “mania.”

While the dominant indexical relation is that of symptom to cause, it would not be foreign to the spirit of Peirce to complicate the analysis with reversible directionality wherein windigo metamorphosis is index and symptoms are indexed objects. Such is attested in cases where symptoms are causes whose neutralization—melting frozen viscera with warm liquids, inducing the subject to eat—reverse the metamorphosis.

Consider dream experience of the type that stereotypically predetermines transformation. The canonical motif is of eating meat later revealed to be (what humans know as) human flesh. From the 1820s in Saskatchewan:

I was invited by the North [spirit] to partake in a feast of ducks, the most beautiful I had ever seen and well-cooked—I set to: a stranger by me touched me with his elbow and said, “Eat thou not of that; look into thy dish”; behold that which I had taken for the wing of a duck was the arm of a child! “Heh! what a narrow escape!' said I.

Then he took me into another room and gave me the most excellent meat, the most delicious in appearance I had ever seen. I would not eat--I discovered it was the flesh of indians thus served up to me.

He took me into a 3rd room and gave me tongues: these I also perceived were the Tongues of indians.
“Why refusest thou what I offer thee? is it not good?
“I feel no inclination to eat,” I replied.

Then he took me in a 4th room where fine beautiful hearts were served up and I was desired to eat, but I perceived that it was still the same. I therefore refused. “Then,” said he, “it is well done--thou hast done well.” Heh! Had I unfortunately eaten of this then had I become a cannibal in addition to all my other misfortunes! (Nelson 1988: 90)

The dream is built internally as (at least) three laminated iconic metaphors likening human zoophagy to windigo anthropophagy, human to windigo perceptions of “game,” and prior dream anthropophagy to its future waking practice. Initially immersed in windigo perspective, the subject here providentially breaks out of it, thus obviating dire consequences.

Is Windigo Subjectivity Real?

Asking whether windigos are real might entail the different question of whether windigo subjectivities were or are by Peircian pragmatic criteria real, “whatever you or I or any man or men may think of them to be” (MS 296:18, cf. Parmentier 1994:19). Algonquian accounts report historical persons as self-identified windigos. By the 1950s, these accounts had been ethnologically fashioned into “windigo psychosis,” a “culture-bound syndrome” whose Algonquian sufferers experienced and acted on cannibal desire.

The phrase “windigo subjectivity” stands in hereafter for “windigo psychosis” to refer to experiences of these self-identified windigos. Leaving aside the labeling issue, aficionados everywhere will recall that windigo subjects’ real-ness was heartily dismissed in the 1980s, written off as a composite of Native superstition (with functional value-added: the belief legitimized logistical homicide [c.f. Marano 1982]) and credulous non-Native Orientalism. More broadly, ethnological writing on windigo and like anthropophagous forms was assimilated to a unitary colonial/neocolonial “discourse on cannibalism” (cf. Barker et al. 1998). Inertially reproduced in Waldram’s Revenge of the Windigo (2004), the authoritative-sounding claim that windigo subjectivities were ascribed but never experienced was becoming default wisdom in a post-millennial anthropology increasingly double-shook by alterity.

Then a new generation of Native and non-Native scholars (and one film-maker) revisited the question, leveling critical postcolonial gazes on skeptics’ summary dismissal both of all Algonquian witness and also of corroborative non-Native descriptions of persons “medicalized” as windigos in Euro-Canadian missions, jails, and psychiatric institutions (Carlson 2009, 2011; Chabot 2010; Smallman 2010, 2014; Tizya 2012). Conclusions that some self-identified windigo subjects existed historically approximate more closely than their contraries to the status of Peircian “final interpretant,” the interpretive result “toward which the actual tends” (SS.111[1909]).

Windigo Subjectivities as Interpretants

Windigo subjectivities were surely indexes of windigo metamorphosis schemata but they also suggest Peirce’s abiding concern with relations that signs contract with habitual dispositions to action (cf. Miller 1996). Such biographical-historical subjectivities were effects of windigo cultural schemata qua signs on windigo subjects qua interpreters. These effects—on ideation, disposition, perception and practice--are best identified as varieties of interpretant.

Peirce initially conceptualized effects on interpreters as instances of that subclass of signs in a condition of reciprocal determination of and by their objects: “The Object of a Sign may be something to be created by the sign.” For example, the command “Ground arms!” has for its object the “immediately subsequent action of the soldiers so far as it is affected by the molition expressed in the command” (CP 8:178[1903], emphasis added).

In later writing, Peirce subsumed signs’ effects within the interpretant. 1906 saw institution, seemingly in the same paragraph, of two classifications; their interrelation is much debated, but each is, of course, triadic (CP 4.536, cf. Short 2004:325). Briefly, emotional and energetic interpretants are respectively effects on interpreters’ dispositions and actions, and logical interpretants are meanings. The second triad comprises immediate interpretability, dynamic effects on interpreters, and final interpretants “to which every Interpreter is destined to come if the Sign is sufficiently considered” (SS. 111[1909]). In c. 1906, the object of “Ground arms!” had become “the will of the officer that the butts of the muskets be brought down to the ground” while the emotional interpretant was the addressees’ compliant obedience and the energetic interpretant the ensuing physical actions and effects (CP 5.473,475). Compare 1909 where the object is “rightful authority” and the interpretant [species unspecified] the subalterns’ mental compliance and physical reaction [MS 634]. The same year the dynamic interpretant was “the thump of the muskets on the Ground or rather it is the Act of their Minds” (CP 8.315).

If “logical interpretants” were always and exclusively general concepts, Peirce characterized “ultimate logical interpretants” of concepts as habits formed in relation to them. Said Peirce, “The deliberately formed self-analyzed habit--self-analyzing because formed by the aid of analysis of the exercises that nourished it—is the living definition, the veritable and final logical interpretant.” Peirce characterized this entity as the changes it induces in “tendencies toward action” and in “habit[s] of acting in a given way” (CP 5.491).

The elements composing windigo subjectivities were of heterogeneous character. Rather than distributing them across Peirce’s interpretant troikas, it suffices here to note their non-fit with the “ultimate logical interpretant.” The latter for Peirce names habits predicated on signs increasingly mimetic of the Real, while windigo subjectivities are based on what he called “indubitable,” i.e., critically unexamined, beliefs and inferences (cf. EP 2:26,346-348[1905]).

Semiotic Mediation and Windigo Subjectivity

Given the pathos and hilarity of earlier nutritional, psychoanalytic, and functionalist “explanations” of windigo subjectivities, a question is how semiotic mediation might serve better. As Teicher (1960) argued long ago, we have here events of “beliefs” engendering some extraordinary “behaviors.” For some Algonquian individuals, anthropophagous ideation could be the indexical icon pointing temporally backward to prior cannibal dream events and forward to like future practice. For Peirce, such ideation would be an indexical icon of conventional beliefs in engendering dreams and engendered metamorphoses, thus, in this respect, a symbol of these beliefs as well. For Peirce, of course, nothing precluded behavior predicated on “indubitable” false propositions.

Windigo subjectivities were mediated by signs whose objects were formulable as dicent symbols (propositions) of the type “cannibal ideation indexes prior predetermining cannibal dream” or “cannibal ideation indexes future cannibal practice.” Interpretants comprising Windigo subjectivities were biographical-experiential replicas of these schemata, more specifically, icons of stereotyped symptoms (anthropophagous ideation, and the like) stipulated in them. Such individuals’ experiences were likenesses of what people were stereotypically supposed to experience when ”going windigo,” and “windigo subjectivity” refers here precisely to lived sensations mimetic of these stereotypes. The symptoms could be both self-ascribed and ascribed by others, and subjects’ reported speech refers to cannibal desire, internal freezing, and theriomorphic perception of relatives. These symptom-icons were sometimes accompanied by threats and warnings or—poignantly--requests for restraint or execution. Any of these behaviors might have compelled Peirce’s agreement with fur trader Alexander the Henry the Elder in the 1790s that these “distressing object[s] appeared to verify the doctrine” [of metamorphosis] (Henry 1901: 200). Progression in metamorphosis coincided with windigos’ increased control over human victims and the latters’ decreased capacity for resistance. In consequence, the signs effectuated in windigo subjects’ companions such interpretants as attempted cure and, less commonly, execution.

Given Algonquian windigo schemata, it is unsurprising that some persons might fear onset of anthropophagous desire and metamorphosis. It is entirely another project to understand how these schemata might engender experience of such desires. The difficulty increases when we pass from interpretants of fearing or experiencing cannibal desire to “habits” of enacting it. Through processes not yet elucidated, these beliefs seemingly induced in some few biographically or temperamentally predisposed persons corresponding dispositions and conducts. How might semiotic mediation get us from stereotypic dicent symbols to anthrophagous ideation as experiential replica-interpretant--and thence, perhaps, to more extreme performance?

There are, radically to understate, many rooms in Peirce’s mansion(s) wherein to look for answers—which answers abide here as important desiderata. Any principled approach would need to focus on Peircian approaches to self and subjectivity (Singer 1984, 1989; Colapietro 1989) and semiotic takes on these more inclusively (cf. Urban and Lee 1989, Mertz 2007). Whatever the denotational infelicities of “psychosis,” Algonquian people themselves speak of windigos as “insane” (note mootha waaskamiithihtam ‘it’s not sane’ as characteristically understated description), and semiotic mediation of “insanity” is potentially germane. Clinical lycanthropy continues alive and well in modernity (c.f. Keck 1988 et al., 1988, Larner 2010, Garlipp et al., 2004), although likeness to windigo subjectivity should not be exaggerated (notwithstanding French Canadian windigo-loup-garou hybrids on which see Podruchny 2004). There exist, however, strong grounds for rejecting claims that windigo subjectivities were necessarily interactive with psychological states commensurable with those enumerated in the Diagnostic and Statistical Manual of Mental Disorders V or like compendia (cf. Carlson 2009). There is here an obscure, esoteric, and controversial domain perhaps most appropriately investigated by members of Canada’s Native Mental Health Association.

The Algonquian doctrine that famine cannibals thereafter acquired preference for human flesh resonates with Pierce’s focus on experientially revised, purposively acquired habits. We are here, however, on evidentiary thin Ice, so to speak. The idea that famine cannibals went on to turn necessity into virtue is not borne out in known historical cases (neither was it an exceptionless Algonquian principle). This is not to say that it never happened. But the single windigo subject whom Natives and non-Native accounts convergently described as killing and eating people in a non-famine context lacked such a history (cf. Thomson 1984 on Swift Runner). This is probably among the reasons Canadian authorities murdered him by hanging. The rarity of such crimes (with or without prior cannibal history) is suggested by non-Native and some Native accounts.

As to etiology, Rock Cree consultants most commonly refer such matters to dream experience. In their discourse, waking events may be indexical icons of prior dream events within which waking event outcomes “have already happened.” Subjects’ cannibal ideation might index a prior dream experience, remembered or not. Inversely, memories of such dream mentation could be anticipatory indexical signs of future experience of cannibal desire. The principal earlier ethnographic corroboration is Kohl on Ojibwe in the mid-19th century. Kohl likened interpretant-effects of windigo belief and of literary art: “It is just like ‘The Sorrows of Werther.’ First there is a Werther in real life, whom the poets render celebrated, and at last the nation is inoculated with Werthers” (Kohl 1985:358).

They begin to dream of them, and these dreams, here and there, degenerate to such a point that a man is gained over to the idea that he is fated to be a windigo. Such dreams vary greatly. At times a man will merely dream that he must kill so many persons during his life; another dreams that he must also devour them; and as these strange beings believe in their dreams as they do in their stars, they act in accordance with their gloomy suggestions. (1985:357-58)

The windigo-subjectivity-as-enacted-dream explanation wants qualification. Algonquians may, indeed, sometimes construe failure to enact dream directives as entailing dire misfortunes. And windigo subjects--hailed, interpellated, and “commanded” by windigo dream-signs--might experience fatalistic resignation to the inevitability of the indexed metamorphosis. But the people did not, of course, regularly enact dream directives contrary to foundational principles of sociality and morality. There is here more in question than reading cannibal desire as an interpretant of “ground arms”-like dreams-that-must-be-obeyed.

There exists, in any event, congruency of Algonquian schemata with Peirce’s take on the semiotic self. Peirce’s held that “man's circle of society (however widely or narrowly this phrase may be understood), is a sort of loosely compacted person, in some respects of higher rank than the person of an individual organism” (CP 5.421). Amongst much else, Peirce professed individuals’ dialogic constitution in interaction with consociates (cf. Singer 1984:57). Windigo subjectivities were anticipatory of incipient transformation, and this accords sinister inflection to Peirce’s dynamic conception of the individual qua sign: “His thoughts are what he is ‘saying to himself, that is, saying to that other self that is just coming into life in the flow of time” (CP 5.421, emphasis added).

Compositions and Laminations

There remains as another desideratum examination of the semiotic blends and laminations through which windigo schemata signify. As concerns the “secondness” troika of icon, index, and symbol, Peirce’s famously dwelt upon their contrasting dependencies relative to objects and interpretants. All must be interpreted if they are to function as signs, but icons and indexes possess semiotic potential (likeness and contiguity, respectively) independently and irrespective of interpretation. In contrast, as Peirce explained, the symbol “represents its object only by being represented to represent it by the interpretants it determines” (MS 599.43). The symbol “signifies what it does only by virtue of its being understood to have that signification” (CP 2.304). Thus the “arbitrariness” of Ojibwe wiindigoo and its cognates as rhematic symbols (nouns)--albeit qualified by descent from Proto-Algonquian *wiintekoowa ‘owl’ whose strigine denotata have cannibal proclivities (Dahlstrom 2003). In addition to conventionality of ground, both the sign and object of wiindigoo exhibit the generality which Peirce, degenerate exceptions aside, built into symbolicity (CP 2.293).

Lexical sign-to-object relation aside, what of different Windigo schemata as reversible signs and objects of one another? Here we find laminations of proximal symptoms on distal causes, and of both these as signs determining windigo subjectivity as interpretant. Consider, for example, the windigo’s frozen heart which attaches the anatomical locus of good and evil (mithotiihii- ‘s.o. has a good heart’, macitiihii- ‘s.o. has an evil heart; [cf. miimiithiitiihiisiw ‘hairy heart being’ or myth age windigo]) to the indexical series composed by the ice and the northern direction (and their associated beings), winter, freezing temperatures, and famine events. Here, the schemata exemplify Peirce’s conclusions that the most felicitous signs are those in which symbolic, indexical, and iconic modalities are blended in approximately equal proportion (in Jakobson 1980).

Peirce’s other two troikas are ethnologically under-celebrated. In the firstness sphere, some glancing use was made above of the qualisign, legisign, and sinsign triad. As concerns thirdness, the rheme, dicent, and argument troika will clarify analyses particularly of such windigo schemata and subjectivities as exhibit disparities between the “actual” sign-object relation and its representation “for its interpretant” (cf. Parmentier 1994 on “upshifting” and “downshifting”). Such disjunctures may evoke differences between animist and modernist ontologies, but also blendings constitutive of the sign’s signifying capacities. Consider, for example, signs whose interpretants know them as being of entirely different type. Of the ten set out in the “triadic classification” of 1903 (CP 2.254-63), Peirce noted of [8] rhematic symbolic legisigns (e.g., common nouns) that their interpretants “often” represented them as indexes [6] (e.g. pronouns), or as icons [5] (e.g. blueprints) and that they partake in small measure of both.

Semiotic and Windigo Perspectivism(s)

A foxy odor is a sign of danger to the rabbit but of dinner to the cougar.
                                                                    T.L Short 2004: 235

Windigos share with animism and perspectivism scant acquaintance with (Peircian) pragmatism and semiotic. Below a frighteningly skeletal sketch of perspectivism serves as context for remarks on the semiotics of its specifically windigo-related manifestations. Left largely unaddressed is the encompassing project of a semiotic of perspectivism.

Animism and Perspectivism

As elucidated (not homogeneously) by Viveiros de Castro (1998, 2009, 2012) and by Descola (2005, 2010, 2011, 2013, 2014a) animist perspectivism ascribes a generalized spiritual endowment ( Descola’s “interiority”) and particularized bodily habitus and hexis (“exteriority”) across different human and nonhuman natural kinds or “species.” I dispense here, for the nonce, with the venerable debate on the lexical appropriateness of “anthropomorphism.” Each kind has anthropomorphic self-experience, albeit particularized by kind-specific attributes, and each commonly experiences other kinds as nonhuman objects. Each animal kind’s reflexivity, for example, thus converges, in variable degree, upon a cultural “human-ness” imperceptible to other kinds. (In Australian varieties, however animal kinds may have reciprocal inter-kind experiences of one another’s anthropomorphic subjectivity [cf. Sahlins 2014]). Viveiros de Castro’s phrase “cosmological deixis” registers likeness of this "exchange" of human subjectivities across kinds to discursive “exchange” of 1st and 2nd (or 1st and 3rd? ) pronominals in person deixis (1998:476-77). Conjoined with animism in Subarctic, Arctic, and Amazonian (and other) terrains, perspectivism also appears in societies of contrasting ontological kind (naturalism, totemism, analogism) in Descola’s quadripartite scheme (Descola 2005, 2013, 2014:297-98).

If human kinds model a generalized interiority, nonhuman experience of them commonly appears in one or the other of two asymmetric modes (cf. Descola 2005:198-99, 2013). Members of non-human kinds either know humans qua humans or as members of one or another different nonhuman kind: Anacondas may know humans as peccaries (as with sharks’ imputed experience of wet-suited surfers as seals), while peccaries know humans as anacondas. In the first case, human kinds uniquely occupy a space in which subjects’ reflexive knowledge coincides with others’ objective experience of them. In the second, humans join other kinds whose shared anthropomorphic interiority may be only reflexively sensible. But humans then vanish entirely as an individuated objective kind upon which different nonhumans might project convergent experience. Jaguar and anacondas may actualize commensurable experience of peccaries but without distinction between those with human and those with peccary reflexivities.1

Semiotics and Perspectivism

Literature on animist perspectivism has been characterized as “neo-structuralist,” and any proposed “semiotics of perspectivism” invokes debates on assorted antipathies and complementarities between Peircian semiotics and semiology-structuralism. It also invokes possible affinities of semiotic mediation with a relativist constructivism with which literature on animist ontology is in explicit disagreement (Descola 2005:345-346, 2013:250-251, 2014a:272-273). Transduction or synthesis would require moving beyond received Saussure vs. Peirce oppositions in directions pioneered by Roman Jakobson and Michael Silverstein.  Spins on semiotics as redemptive antidote to ahistorical, agency-proof, context-free, and materiality-deficient structuralism(s), not entirely untrue, are both oversimplified and overdone. There was, for example, the Prague School. Consider Jakobson’s indisputably structural anchoring of distinctive features in acoustic materiality.

Principles composing animist perspectivism are construable both as objects of other signs and as themselves composing a complex sign with multiple objects. Characteristic signs of perspectivism as object are remembered dream mentation in which nonhuman interlocutors have human anatomy and disposition, and literary motifs where subjects of different kinds have simultaneous contrasting sensory experiences of the “same” object (especially as body part vs. artifact) or where nonhumans transit reversibly between distinct chronotopes in which human and animal characters are foregrounded. Descola (2010) examines visual signs of perspectivism in fine detail.

Perspectivist ontology is expressed in animist discourse (rarely) and scholarly writing (often) as Peircian dicent symbols, e.g., propositions representing either particular instance or general law.  The proposition is “dicent” in Peircian idiom because its interpretant represents the sign as an index connected to and influenced by its object (whatever the presence/absence of such connection). The proposition is the composite of indexical subject and iconic predicate, and its complex object is the subject’s possession of the predicate object’s qualities (CP 2.262).  It would be possible to analyze in these terms such clause-level propositions as “Different exteriorities actualize distinct worlds.” One can also make an assemblage of such animism and perspectivism-characterizing propositions (universalized interiority and particularized exteriorities, subjective“ humanness and objective non-humanness, transient participations in inter-kind humanness, reversible inhabitance by different human and nonhuman kinds of predator-prey roles) and ask what such an assemblage, qua complex sign, has as object.  One folk-anthropological answer is monistic oneness of humans and nonhumans.  A less remarkable but plausible object is intuition that likeness across kinds is greater than it seems. The convergence with Berkeley’s immaterialism noted by Descola (2005:390, 2013: 284) is another possible object.

Some recent writing on animism tries to reconceive or displace structuralist binaries of interiority/exteriority, soul/body, predator/prey, and subject/object. Viveiros de Castro introduced a soul/body distinction based on reversibility: “The invisible dimension of the invisible dimension is the visible one, the soul of the soul is the body” (2001:42). Thereafter, “reversibility” (in the meaning that “reversibles are opposites that contain themselves” [Willerslev 2007:521]) together with recursivity, and Deleuzian multiplicity(ies) have acquired increasing salience (cf. Fausto 2007, 2011; Willerslev 2011). These currents have implications for semiotic construals of perspectivism.

Human Game Animals

If in some Amazonian societies, spirits and predator animals (jaguars, anacondas) know humans only as prey animals, Subarctic Algonquian ontologies exhibit the other asymmetry wherein nonhuman kinds know humans as humans. But to this the windigo is the prominent exception, reproducing Amazonian themes of spirits whose perspective makes of humans just those animals prominent in human diets. A signature element in windigo schemata is visual, auditory, and olfactory experience of humans as game animals. Windigo perspectivism is described most authoritatively by the sociologist Pynchon:

Get the picture,” he had told Grossmann that night, over mugs of Wurtzburger. “Altered perception. Simultaneously, all over God knows how many square miles, hundreds, thousands of these Indians are looking at each other out of the corner of their eye and not seeing wives or husbands or little children at all. What they see is big fat juicy beavers. And these Indians are hungry, Grossmann. I mean, my gawd. A big mass psychosis. As far as the eye can reach” – he gestured dramatically – “Beavers. Succulent, juicy, fat.” (Pynchon 1959)2

This point of view is shared by myth-era windigo-kind collectives (who, by the way, ate not only people but also one other) with the solitary metamorphosized windigos known historically. As for the former, wiihcikoosisak ‘small windigos’ knew humans as “moose” in address and reference alike (Brightman 1988:135). As for the latter, consider the reported speech of an executed windigo subject in 1899: “He said that his children appeared to him as young moose” and asked "How would it do if I should eat my little ones, and especially their noses?" (Teicher 1960). Windigo subjects’ reported speech also commonly describes reflexive cognizance of disjunction between their own and others’ perceptions, and, also reversible shifts between the two.

It rests mysterious whether fully-transformed windigos retain such meta-awareness, know difference between human and nonhuman, or recognize humans qua humans. Limited evidence suggests that they perceive animals as humans do (vs., say, perceiving them as human) while taking, of course, no dietary interest in them. In one narrative (somewhere in the upcoming chapter), fugitive humans strategically hide beneath a moose effigy, confident of the pursuing windigos’ indifference to it. The indexes of difference between edible human “moose” and inedible nonhuman “moose” wants elucidation. (But this, of course, is the point.) It is no accident that windigos commonly knew humans as herbivorous beaver or moose but seldom as bears whose omnivory sometimes inverted the status quo.

The privileged exception to this generality emerges in shared ontological spaces of animism and totemism (Descola 2005:232-240, 2013:165-171, Sahlins 2014). Some boreal forest Algonquians, their animism notwithstanding, assert consubstantiality (sometimes based in ancestral “descent”) of patrilineal clan members with individuals of their eponymous totem species, doctrines commensurable with those defining Australian totemism in Descola’s classification. Thus in Southwestern Ojibwe society, windigos’ theriomorphic perceptions of human prey might be “speciated” by the latters’ participations in their totem species:

This is what my grandmother said: “He who is a windigo sees the other Indians as their totems. He sees anybody who has a bear for his totem as a bear, and so he kills and eats him, and so with someone who has the deer as his totem. If anyone has a beaver as his totem, that’s how he sees him and so he kills and eats him.” (Kegg 1990:15)

Bear, deer, and beaver are Ojibwe clan totems; odoodeman does not here describe individuals’ dream beings. Note that desubjectified windigos own erstwhile totem affiliations do not mediate their perceptions.3

In capacity as “jaguar gone north,” the windigo has another Amazonian resonance. Taylor (1993, cf Viveiros de Castro 1998:483-84) has analyzed encounters wherein, whether through ingestion or discourse, nonhumans assimilate human interlocutors to their kind. Windigo victims become neither windigos nor animals but are subject to controls over their will and action requiring extraordinary effort to resist. “Staying alive was made less difficult for a Wetiko by the fact that most people were rendered helpless just by sighting it” (Merasty 1974). In this measure, victims participate in windigos’ perspective on them as prey.

Animistic Non-Subject

For many Algonquians, as elsewhere in animistic archipelagos, each nonhuman kind has reflexive “humanness” of which humans may acquire knowledge in dreams. If windigos know humans as nonhuman prey kinds, symmetry would suggest that they also know reflexive humanness. Au contraire. The windigo claims exemption from the anthropomorphic subjectivity animist perspectivism would generalize, in principle, across nonhuman kinds. If spirits, animals, botanicals, and even manufactures experience themselves and others of their kind as enculturated, discursive, and enskilled, the signature feature of the metamorphosized windigo is its deculturated and aphasic state.4 Unique among nonhuman kinds, it is ascribed no cultured subjectivity. Here sartorial style (“Windigo is reported to be an evil spirit, a huge individual, who goes naked in the bush and who eats Indians” [Davidson 1928:263]) and culinary preference (raw human flesh) combine with a certain technological incapacity. Merasty’s (1974) is the best discourse on the subject:

In general, the ability of the Wetiko was below that of a normal person. Because it could not build a canoe, it could be stopped by a lake. Because it could not make a shelter, it had to find one ready-made, The Wetiko was an animal that was less keen and fit for survival than the other residents of the forests in which it roamed. (Merasty 1974:8)2a

Merasty’s “animal” descriptor is perhaps metaphoric and is, in any case, unsubstantiated by evidence that windigos acquire subjectivities of predator animal kinds--which would, in any case, parallel the anthropomorphic subjectivity the windigo has left behind. If the windigo neither knows itself as human nor assimilates to an animal kind, how does it know itself?

The wetiko, once a man hunting animals, became an animal and a hunter of people, without knowing what had happened. It did not remember that it had once been a normal human bring and probably could not recognize the similarity between itself and other human forms. In the eyes of the Wetiko, humans were game. (Merasty 1974:14, emph. added)

As refracted obliquely through human experience of them, windigo awareness is a mélange of single-minded ravenousness, aggression, and stupidity. Their at best distressed subjectivity accords with a singular position in the thematics of metamorphosis. While other differentiated kinds retain the interiority shared in common by their distinct myth-time prototypes, windigos are transformed erstwhile members of the kind they now feed upon. Windigos thus live entirely outside of animism’s universalized, anthropomorphic, and specifically enculturated interiority. As befits their non-subject status, windigos do not accord to other kinds the human-like reflexivity such kinds accord one another across the animistic archipelago.

Zoophagy and Anthropophagy

Moral and practical dilemmas posed by zoophagy for animists were noted for Subarctic Athapaskans by McClellan in long-ago 1970, discussed in the Amazonian context by Viveiros de Castro in 1992 (c.f. 1998:481) and then raised in the Algonquian Subarctic where co-existing positive and negative models of hunter-prey relations exist (Tanner 1978, Brightman 1993). In the later context, a question arises: “If killing and eating index exploitation and if animals are beings like humans, what differentiates the [Cree] hunter from the sorcerer and the witiko?” (Brightman 1993:203). The topic continues prominent in Amazonia from whence Viveiros de Castro (1998:481) describes “the phantom of cannibalism” and Fausto (2007:498) asks whether “food consumption [might] necessarily slip into cannibalism.” Descola (2005:36-7, 391-96) has addressed animistic cannibalism and compensatory practices comparatively on global ethnographic scale.

For hunter-gatherer societies in particular, some authorities either ignore negative modes of human-animal sociality (Bird-David 1992, Ingold 2000:69) or are incapable of differentiating them from co-existing positive counterparts (Nadasdy 2007). In contrast, Descola (2005: 426-39, 2013: 309-321), distinguishes unconditional partage (sharing), reciprocated échange (exchange), and exploitative prédation (predation) as categories of human-nonhuman relationality present in animist systems. Algonquian “positive” and “negative” modes distribute across these with the expected correspondences.

Descola’s theoretical apparatus allows for co-presence of the three modes (and others) in single societies, one hierarchically dominant and the others ancillary. This notwithstanding, he asserts that Subarctic Algonquians know human-animal relations exclusively as “sharing” and are thus “liberated from the dangers of predation and the constraints of exchange” (Ibid:489; 2013: 358).

[Algonquians] concoivent le milieu subarctique … comme un environment bienveillant et peuplé d’entities attentives aux besoins des hommes. C’est toujours par l’effet d’un sentiment de generosite que le gibier se livre au chasseur: mu par la compassion pour les humains tenaillés par la faim, il fait don de son envelope charnelle sans attendre de compensation. (Ibid:489-490)
[Algonquians] regard the subarctic region ... as a benevolent environment that is inhabited by entities that are attentive to the needs of humans. It is always out of a feeling of generosity that a hunted animal delivers itself up to the hunter. Moved by a compassion for humans in the grip of hunger, it presents him with its carnal envelope, as a gift, without expected any compensation. (Descola 2013: 359)

Having pronounced predation absent in Algonquian human-animal relations, Descola finds its solitary expression in human-windigo relations:

il est vrai que la hantise de la prédation n’est pas complètement absentes des groupes alqonquins septentrionaux; elle y prend la figure du Windigo (ou Wiitiko), un monster cannibal a forme humaine qui inspire un grand effroi aux Indiens. (Ibid:578)

Admittedly, the obsession with predation is not totally absent among the northern Algonquian groups. Here it is represented by the Windigo (or Wiitiko), a cannibalistic monster in human form that terrorizes the Indians. (Descola 2013: 426n30)

Given the scope of Descola’s theoretical project and the literally planetary distribution of his ethnographic sources, some classificatory problems are expectable. It is, therefore, unsurprising that the hierarchical priority of Algonquian sharing (more usually reciprocal exchange) is here mischaracterized as an exclusive presence. Exchange and sharing are, indeed, the primary forms. Consider the (Woods) Cree expression ta-asamisowan ’it [animal] will feed itself to you [hunter].’ But the positive mode of human-animal relation objectified in sacrifice and honorific song is in ironic juxtaposition with the discourse and practice of hunting as exploitative sorcery and violence. As schema of hunting, Algonquian exchange euphemizes Algonquian predation—although this ideological property hardly exhausts the former’s character and qualities.5

Descola is entirely correct on windigo as Algonquian virtuoso of prédation, but the picture is completed by composing the windigo together with its doppelgänger, the predatory human hunter. In terms of contrasts between hyper-subjectification and desubjectification of human and nonhuman meats (Viveiros de Castro 1992:290-93, Fausto 2007), the windigo, unlike the human hunter (some exceptions aside), consumes its prey as raw subject rather than cooked object. (Algonquian hunters, of course, continue intersubjective communication, in musical and other registers, with the animals whose flesh they desubjectify and eat). Unlike sharing-and-exchanging human hunters, windigos make over no sacrifice to human victims, sing them no honorific song, accord them no respectful mortuary deposition. Neither do victims “come to be like human” in windigo dreams. No doctrine exists that humans love or pity the windigo, voluntarily surrendering their fleshly bodies as a mode of sharing or exchange. Like the predatory hunter, the windigo deploys sorcery and violence against involuntary victims. Windigo predation and the predatory hunting mode are parallel metadiscourses on hunting schema based on sharing and exchange.

Returning climactically to Peirce and semiotics, there exists in windigo schemata some warrant for characterizing perspectivism as a semiotic blend in which iconicity is hierarchically prominent.  Peirce distinguished three varieties of icon. In the image, the sign’s quality resembles the object’s quality and in the diagram, relations between the sign’s parts  resemble relations between the object’s parts. Not faulting these, one might proceed further to Peirce’s stately characterization of metaphors which “represent the representative character of a representamen [sign] by representing a parallelism in something else”  (CP 2.77).  Peirce’s Zen brevity has grown a large, lively, and heterogeneous literature on what he intended (cf. Lottmann 2012:537 for sources and a classification thereof).  I experiment here with Lottmann’s reading which characterizes metaphor as meta-iconic, one sign taking another as object.  The metaphor is composed triadically. The metaphoric sign represents the representative quality of a different sign (as “vehicle”) for “something else” (as “tenor”) than its own conventional object.  The metaphoric object itself is internally composed of this different sign, its novel object, and a discerned parallelism as interpretant.  The metaphoric interpretant overall is that revelation of “unexpected truth” Peirce (2.279) ascribed to iconicity.

Looking back across intervening sections, recall the Cree who described his ill-omened dream to George Nelson in the 1820s:  “Behold that which I had taken for the wing of a duck was the arm of a child!”  A canonical perspectivist motif juxtaposes contrasting simultaneous experiences of the “same” entity by different subjects, but here a single subject moves successively (and reversibly) between such actualizations. The metaphoric sign is the utterance describing and contrasting these perceptions.   Qua sign, this utterance has as object the representational quality of the sign “duck wing” not for its conventional (ornithological) object but for the novel object “child’s arm” which the metaphor assigns to it.  The parallelism originated by the metaphoric sign includes but overruns the anatomical imagery.  Initial perception of the object as edible meat is an icon of the windigo “point of view” (“POV”) in which the subject intitially participates and which actualizes (self-identified) human beings (or parts thereof) as animal resources.  The revelatory discovery of the object as human flesh is an icon of the “fellow-human POV” which actualizes humans as humans and into which the subject opportunely shifts.  In short, “windigo POV” experiences as [ + edible] “duck wing” what “fellow-human POV” actualizes differently as [ - edible]  “child’s arm” and distinguishes from [ + edible] “duck wing.”  Of relevance to the dialectic of prédation and échange is a second iconic metaphor in which the sign  “windigo POV” (on humans as edible animals) acquires as its new object the “human “POV” both on inedible fellow-human and edible ducks. In earlier structuralist formulae: windigo : human  ::  human : animal.  The parallelism represented in the sign’s interpretant includes the disparity in each case between the prey’s anthropomorphic subjectivity and the predators’ actualization of it as non-human.  Windigos’ incapacity to know their victims as human stands for human objectification of animal subjects. Moral appraisal of windigo predation stands for conjectures about animals’ assessments of predation in the quotidian hunt. Since windigos no longer participate in the interiority animism generalizes across human and nonhuman kinds, their meals are the least cannibalistic of any in the animistic archipelago.6

The metaphor in which windigo perspective on Algonquians as sign acquires Algonquian perspective on game animals as object is commonly in absentia but becomes explicit in the case of bears. Most Crees I know relish bear meat, but a minority pass on it, specifically on grounds of anthropomorphic qualities. When asked if he ate bear meat, a Cree replied indignantly, “Do you think I am a cannibal?" (Cockburn 1984:44). Algonquian interpretants of signs of windigo perspectivism and of negative hunting practices alike might be imagined as commensurable, in some respects, with German counterparts evoked by Albrech Durer’s 1504 painting of the “Head of a Stag“ with the arrow shaft lodged between its eyes.


The paper benefited from discussions with Marshall Sahlins, Philippe Descola, Marie Mauzé, and Frédéric Laugrand. Maria Lepowsky made inspired design suggestions conducive to ending the piece with a bang instead of a footnote. Dale Pesmen reminded me of Zepho Beck and introduced me to indispensable and obscure sources on 18th century Philadelphian castoranthropy. Charles Nims took up a musical philosopher’s point of view on Peirce’s nomenclatural practices. Paul Manning was an exemplary model of shrewd editorial intervention and serene patience.


NOTE: References to Charles Sanders Peirce’s writings, published and not, occur with the usual bibliographic formulae.

Atwood, Margaret. 1996. Strange things: The malevolent North in Canadian literature. Oxford: Oxford University Press.

Avramescu, Cǎtǎlin. 2009. An intellectual history of cannibalism. Princeton: Princeton University Press.

Barker, Francis, et al., eds. 1998. Cannibalism and the colonial world. Cambridge: Cambridge University Press.

Bird-David, Nurit. 1990. The giving environment: Another perspective on the economic system of gatherer-hunters. Current Anthropology 31:183-196.

Brightman, Robert. 1988. The Windigo in the material world. Ethnohistory 35(4):337-379.

____. 1989.  Acaðōhkīwina and ācimōwina: Traditional narratives of the Rock Cree Indians. Regina: Canadian Plains Research Center.

____. 1993. Grateful Prey: Rock Cree human-animal relations. Chicago: University of Chicago Press.

Buckser, Andrew and Susan. 1999. New research on windigo psychosis. Anthropology News 40(2):31.

Carlson, Nathan D. 2009. Reviving Witiko (Windigo): An ethnohistory of ‘cannibal monsters’ in the Athabasca District of Northern Alberta, 1878–1910. Ethnohistory 56 (2009): 355-94.

____. 2011. Searching for Catherine Auger: The Forgotten wife of the Wihtikow (Windigo). Recollecting: Lives of Aboriginal women of the Canadian Northwest and Borderlands. Sarah Carter and Patricia Alice McCormack, eds. Pp. 225-44. Edmonton: Athabasca Press.

Chabot, Cecil. 2010. Witiko possession and starvation cannibalism among the James Bay Cree: Monstrosity or madness? Creating humanity, discovering monstrosity: Myths & metaphors of enduring evil. Elizabeth Nelson, Jillian Burcar, Hannah Priest, eds. Pp.3-15. Inter-Disciplinary Press.

Cockburn, R. H., ed. 1985. "Like words of fire": Lore of the Woodland Cree from the journals of R. H. Downes. The Beaver 315:37-45.

Colapietro, Vincent M. 1989. Peirce's approach to the self: A semiotic perspective on human subjectivity. Albany: SUNY Press.

Dahlstrom, Amy. 2003. Owls and cannibals revisited: Traces of windigo features in Meskwaki texts. Papers of the 34th Algonquian Conference. H. Christoph Wolfart, ed. Pp. 81-114. Winnipeg: University of Manitoba.

Davidson, D. S. 1928. Some Tete de Boules tales. Journal of American Folklore 41:262-74.

Descola, Philippe. 2005. Par-delà nature et culture. Paris: Editions Gallimard.

____.   2010.  Un monde animé: Présentation. La Fabrique des images. Visions du monde et formes de la représentation. Philippe Descola, ed. Pp. 23-38. Paris, Somogy-musée du quai Branly.

____.2011. L'écologie des autres: L'anthropologie et la question de la nature. Paris: Éditions Quae, coll.

____.2013. Beyond Nature and Culture. Translation of Descola 2005. Translated by Janet Lloyd. Chicago: University of Chicago Press.

____.2014a. Modes of being and forms of predication. Hau: Journal of ethnographic theory 4 (1): 271-280.

____.2014b. The grid and the tree: Reply to Marshall Sahlins’ comment. Hau: Journal of ethnographic theory 4 (1): 295–300.

Fausto, Carlos. 2007. Feasting on people: Eating animals and humans in Amazonia. Current Anthropology, Vol. 48(4) : 497-530

____. 2011. Le masque de l’animisme: Chimères et poupées russes en Amérique indigene. Gradhiva 13: 49-75.

Doughty, Roy Dean. 2000. Week Ten: Friday’s Rock, Moby Polyphemous.

Foti, Claudio. 2013. Windigo. Mito e leggenda. Rome: Editions Parallelo45.

Garlipp P, et al. 2004. Lycanthropy—psychopathological and psychodynamical aspects. Acta Psychiatrica Scandinavica 109 (1): 19–22

Henry the Elder, Alexander. 1901 [1809]. Travels & adventures in Canada and the Indian Territories between the years 1760 and 1776. Parts 1–2. Toronto: George N. Morang.

Ingold, Tim. 2000. From trust to domination: An alternative history of human animal relations. The perception of the environment: Essays on livelihood, dwelling and skill. London: Routledge.

Jakobson, Roman. 1980. The framework of language. Ann Arbor: Michigan Studies in the Humanities

Keck P. E., et al. 1988. Lycanthropy: Alive and well in the twentieth century. Psychological Medicine 18 (1): 113–20.

Kegg, Maude. 1990. Nookoomis gaa-inaajimotawid: What my grandmother told me. John Nichols, ed. Oshkabewis Native Journal Special Edition 1(2):1-185

Kohl, Johann George. 1985 [1860]. Kitchi-Gami: Life among the Lake Superior Ojibway. Minneapolis: Minnesota Historical Society Press.

Larner, Andrew J 2010. Neurological Signs: Lycanthropy. Advances in Clinical Neurocience and Rehabilitation 10 (4): 50.

Lattmann, Claas. 2012. Icons of novel thought: A new perspective on Peirce's definition of metaphor. Semiotica 192: 535-556.

Lee, Benjamin. 1997. Talking heads: Language, metalanguage, and the semiotics of subjectivity. Durham: Duke University Press.

Levy, Paul. 2013. Dispelling wetiko: Breaking the curse of evil. Berkeley: North Atlantic Books.

Marano, Lou. 1982. Windigo psychosis: The anatomy of an emic-etic confusion. Current Anthropology 23(4):385-412.

McClellan, Catharine. 1970. The girl who married the bear. National Museum of Man Publications in Ethnology 2. Ottawa: National Museums of Canada.

Merasty, Bill. 1974. Comments. The World of wetiko: Tales from the Woodland Cree by Marie Merasty. Saskatoon: Saskatchewan Indian Cultural College.

Mertz, Elizabeth. 2007. Semiotic anthropology. Annual Review of Anthropology 36:337–53.

Miller, Marjorie 1996. Peirce’s conception of habit. Peirce's doctrine of signs: Theory, applications, and connections. Vincent Michael Colapietro and Thomas M. Olshewsky, eds. Pp. 71-78. Berlin: Walter de Gruyter.

Nadasdy, Paul. 2007. The gift in the animal: The ontology of hunting and human-animal sociality. American Ethnologist 34(1):25-43.

Nelson, George. 1988. The orders of the dreamed: George Nelson on Cree and Northern Ojibwa religion and legend. Jennifer Brown and Robert Brightman, eds. Winnipeg: University of Manitoba Press.

Parmentier, Richard. 1994. Pierce divested for non-initiates. Signs in society. Pp. 3-22. Bloomington: Indiana University Press.

Podruchny, Carolyn. 2004. Werewolves and windigos: Narratives of cannibal monsters in French-Canadian Voyageur oral tradition. Ethnohistory 51.4 (2004) 677-700.

Pynchon, Thomas. 1959. Mortality and mercy in Vienna. Epoch 9(4).

____. 1997. Mason & Dixon. New York: Henry Holt.

Sahlins, Marshall. 2014. On the ontological scheme of Beyond nature and Culture. Hau: Journal of ethnographic theory 4 (1): 281-290

Short, T.L. 2004. The development of Peirce's theory of signs. The Cambridge Companion To Peirce. Cheryl Misak, ed. Pp. 214–240. Cambridge: Cambridge University Press.

____.2007. Peirce’s theory of signs. Cambridge: Cambridge University Press.

Singer, Milton. 1984. Signs of the self. Man’s glassy essence: Explorations in semiotic anthropology. Pp. 53-73. Bloomington: Indiana University Press.

____.1989 Pronouns, persons, and the semiotic self. Semiotics, self ,and society. G. Urban and B. Lee, eds. Pp. 229-96. Berlin, New York Mouton de Gruyter.

Skinner, Alanson. 1916. Plains Cree tales. Journal of American Folklore 29(113):341-67.

Smallman, Shawn. 2010. Spirit beings, mental illness and murder: Fur traders and the windigo in Canada’s boreal forest, 1774 to 1935. Ethnohistory 57(4):571-596.

____. 2014. Dangerous spirits: The windigo in myth and history. Victoria B.C.: Heritage House.

Tanner, Adrian. 1979. Bringing home animals: Religious ideology and mode of production of the Mistassini Cree hunters. New York: St. Martin's Press.

Taylor, A.-C. 1993. Des Fantômes stupéfiants: langage et croyance dans la pensée achuar. L'Homme 126/128: 429-47.

Thomson, Colin Argyle. 1984. Swift Runner. Calgary: Detselig Enterprises.

Tizya, Christian. 2012. Cannibal possession: Heart of ice. Eyes Wide Open Films.

Urban, Greg and Benjamin Lee, eds. 1989. Semiotics, self, and society. Berlin, New York:  Mouton de Gruyter.

Viveiros de Castro, Eduardo. 1992. From the enemy’s point of view: Humanity and divinity in an Amazonian society. Chicago: University of Chicago Press.

____.1998. Cosmological deixis and Amerindian perspectivism. Journal of the Royal Anthropological Institute 4:469-488.

____.2001. Gut feelings about Amazonia: Potential affinity and the construction of sociality. Beyond the visible and the material. Laura Rival and Neil Whitehead, eds. Pp. 19–44. Oxford: Oxford University Press.

____. 2009. Métaphysiques cannibals. Lignes d’anthropologie post-structurale. Paris: Presses Universitaires de France.

____. 2012. Cosmological perspectivism in Amazonia and elsewhere. Masterclass Series 1. Manchester: HAU Network of Ethnographic Theory.

Willerslev, Rane. 2007. Comment on “Feasting on people: Eating animals and humans in Amazonia” by Carlos Fausto. Current Anthropology 48(4): 520-21.

____. 2011. Frazer strikes back from the armchair: A new search for the animist soul. Journal of the Royal Anthropological Institute 17(3): 504–526.


1. Amazonian spirit “points of view” may also “merge” what humans distinguish as different nonhuman kinds. There is more to be said on “spirits” and perspectivism in the Algonquian Subarctic. Humans and spirits may project different perspectives on the “same” nonhuman kind. In a Plains Cree myth, a man does bride-service for Thunderbird affines, participating in their (almost) anthropomorphic reflexivity. From his father-in-law, he learns that Thunderbirds know Underwater Panthers as “bears”: "You people [humans] who live below call them [bears] Mici Pisi [Giant Panthers]" (Skinner 1916:359). As for “bears” known to and hunted by humans, these for Thunderbirds are (“classificatory”?) “junior siblings” (Ibid:362).

2. The author here would here seem to be having precocious fun with excesses in earlier “windigo psychosis” interpretations, plausibly as communicated at Cornell University in the 1950s. Note also the same author’s “kastoranthropic” Zepho Beck, unprepossessing alike to “legitimate beavers” and local Indians (Pynchon 1997:619-22). Doughty (2000) provides other examples of castoranthropy, albeit without benefit of references. The minutes of the Quaker Annual Meeting for Philadelphia in the mid 1760s describe visitations by a were-beaver who gnawed the dwellings of evil-doers. “More curious yet, there exists an official civic memorandum from the annals of the corporation in which a were-beaver is implicated in the disappearance of an entire Philadelphia family, as well as that of a prominent local constable.”

3. The passage is relevant to Sahlins’ (2014) proposed inclusion of Descola’s animism, totemism, and analogism as varieties within an encompassing animist ontology, and to Descola’s (2014b) responses.

4. This aphasic-deculturated feature is not shared by myth-age windigo collectives whose members have anthropomorphic self-experience, speech, and culture. The literary motif of the windigo transiently or permanently (re-) enculturated via affinity with human spouses (see Smallman in press) attests reflection on these features, affording also another Amazonian resonance.

5. An Amazonian parallel is Jivaroan euphemization of predation as Dravidian affinal exchange:

La chasse se présente comme l’expression d’une complicité entre desparents par alliance ou le terme ultime, la mise a mort, est occulté par des formules ludiques . . . Mais traiter le gibier comme un affin n’est qu’une tromperie permettant de déguiser la nature foncierement inégalitaire du rapport entre les homes et leur victims animals. (2005:468)

Less clear in the Jivaroan case are parallels to the exchange mode hierarchically dominant in Algonquian discourse.

6. Much remains to be done, semiotically or otherwise. Avramescu’s An intellectual history of cannibalism (2009) has rich comparative potential. The same may be true of Claudio Foti’s Windigo: Mito e leggenda (2013). A semiotic analysis of scholarly windigo “humor” (Buckser and Buckser 1999) and of why ethnologists and Algonquians each sometimes find windigos funny is wanted. So is an account of why the Viking clothing company markets men’s (not women’s) jackets named “windigo” with red and black color options. Literary windigos in Euro-derived genres (cf. Atwood 1996) exfoliate, while the internet has imbued windigos with unprecedented properties of planetary dispersion and metamorphosis. Connections with “otherkin” communities want elucidation. What is the semiotic object of Paul Levy’s book, Dispelling Wetiko: Breaking the Curse of Evil (2011)?