sun, is enough to make a man mad with the thought that he is not a migratory bird. Janet Frame ( 1965:4). All people name and classify animals. The.
Mankind, 12 (1979), pp. 13-27
Animal Metaphors: An Evolutionary Model J
PETER D. DWYER* Mass movements, migrations in time have been made by man pursuing or fleeing from an idea. He settles around the chosen idea. He breeds there. This everlasting movement back and forth in time without the stability and guidance of a visible world, recognised seasons, shared sun, is enough to make a man mad with the thought that he is not a migratory bird. Janet Frame ( 1965:4) All people name and classify animals. The classifications employed are of hierarchial kind; they are taxonomic. They vary between cultures in the range of animal types included, in the ways these types are arranged into more inclusive categories and in hierarchial depth. As well, all people extend the referents of some named animal categories in metaphoric ways. They analogize properties of animal categories to those of social existence. Again there is variation between cultures. Magic and myth, totemic labelling or cultural identity may all be aided by animal metaphors. When we analyse the zoological taxonomic systems used by people we focus upon the supposed denotative functions of names. When we analyse modes of metaphorization we are explicitly concerned with connotative functions. But should we connect or disconnect these two modes of human signification? This is a recurring and controversial question in anthropology. It has been answered variously. For some anthropologists the logical basis for both modes of signification is the same; relationships between certain animals, made explicit in taxonomy, are re-presented in social existence when those animals are used as metaphors for particular configurations of people or of acts (Lkvi-Strauss, 1966). For others, metaphors arise haphazardly or fortuitously and metaphorization is thus in vivid contrast to the ‘protoscientific’ form of * Zoology Department, University of Queensland, St Lucia, Queensland 4067. Ms. received February 1979.
taxonomic systems (Hiatt, 1969; Willis, 1975; Worsley, 1967). Or again, it has been intimated for particular societies that the taxonomic system might constrain the choice of metaphors allowing us to predict a pattern for the latter from a knowledge of the former (Bulmer, 1978; Douglas, 1970; Tambiah, 1969). In this essay I shall show how we may connect the taxonomic and metaphoric domains of human significati0n.l I shall disagree with the opinions that the logical bases of taxonomy and metaphorization are the same, that metaphorization is haphazard or fortuitous and that prescientific taxonomies are protoscientific. I shall generalize the opinion that extant taxonomic systems may constrain metaphoric possibility. My perspective is evolutionary. I shall describe a process of taxonomic elaboration and argue that as it unfolds new and different styles of metaphorization become available while the contexts in which particular styles find expression are restricted. The model I develop is predictive only as regards styles and contexts. It refers to potentialities and constraints that surface as evolution proceeds but it does not and cannot specify results. This assertion is crucial. In evolution of any sort there is, as Gregory Bateson noted, a contrast between ‘continuty of change and discontinuity of the classes which result from change’ ( 1958 : 283). Process and product are distinct yet interact. An adequate description of process merely defines universes of potentialities within domains of realized potentialities (i.e. products) that have gone before. The posture of stasis that prevails in much modern science has often meant that we depict moments of process as though they were products. I n this we can lose sight of both; in what follows I shall try to keep them in cocus.
MANKIND
VOL. 12 No. 1
Modern conceptions of evolution, whether de novo as Wagner implied; they arise conorganic or cultural, are concerned with the textually. And they are subject to accepmanufacture of variety. They describe in- tance or rejection in a public arena. Only determinate trajectories. They usually view where they condition comparable percepquestions ,of ‘direction’ in evolution with tions in fellow humans will they be resome embarrassment or else they dismiss peated, gain acceptance and extend cultural them as teleological. A consideration of meaning. Otherwise, like bad puns, they direction is fundamental to my approach. A will be ignored. retrospective on evolution presents a Metasignifiers express and lend force to sequence of patterns, not merely as variants relations newly perceived in a familiar upon an organizational theme, but of the world. Their form is analogic; they connote organizational themes themselves. Com- relations. Yet they originate within a plexity or the information content of sys- matrix of previously accepted contrasts, that tems increases, heterogeneity arises from is, a classificatory matrix, and in winning homogeneity, energy is channelled more acceptance they reinforce contrast. In efficiently, productivity increases. In retro- analogizing the properties of discrete spect we may recognize a trend through signified components, metasignifiers insisprogressive individuation of ever-changing, tently promote the discreteness of those more intricate, patterns of involvement components. They alter the referents of (Langer, 1967:Ch. 9 ) . No morality of subordinate signifiers by depicting new re‘progress’ is implied here. What is called lations between the components concerned. ‘increase in organization’ must entail con- In doing so they extend cultural meaning. straints upon change and potentialities for Yet, through usage, metasignifiers may change that become manifest and sub- decay; they may come to denote, rather sequently are realized as evolution proceeds. than connote, the relations or (with an eye No goal-seeking interpretation of evolution to taxonomy) the set of components that is necessary. The evolutionary trajectories fostered their origin. They will now be considered in this essay are directional in eminently available as source material for the above sense. later metasignification.2 FIG. 1 gives a taxonomic example. It The evolution of taxonomy depicts a fragment from an hypothesized My central notion for change in the expres- history of taxonomic change for Rofaifo sion of culture is shown below. It concerns people of the Eastern Highlands District of innovation; it modifies and restates Wagner’s Papua New Guinea. The faunal domain in theory of ‘cultural meaning’ in an ecological question comprises ‘insects’ and frogs and setting (1975; see also Barnett, 1953). Its its history is modelled to accord with the referents are not just taxonomic. It is process just described. But it is crude; too analonous to a recent uroDosal concerning simple an exercise. I present it to quickly the &ohtion of bioiogi’cal organizatioi stress that higher taxa (originate as metasignifiers; they originate in analogy and be(Riedl, 1977). come incorporated into lexical systems of signifier + signified perceived ~ o n t r a s t . ~ metasignifier similarity Metasignifiers will exist in either a comsignifier + signifiedJ plementary or contradictory relationship to ? one another. Wagner suggests that a set of Innovati,on upon the meanings of a cul- metasignifiers whose relationship ‘is completure becomes possible where similarity is mentary, since they refer to different and perceived between components of the ex- non-conflicting aspects of the same cultural tant socio-ecological environment, or be- domain . . . , constitutes an idealogy’ tween properties of those components, and ( 1975 :169-70). Ideological domains may is articulated through metasignification. The contrast with others, thus affording possicontext in which metasignification arises is bilities for metasignificatory extension simply the environment of the familiar. The across domains that will reinforce domain possibilities for extension are numerous but contrast while creating new domains. The circumscribed. Metasignifiers do not arise flow of metasignification generates an ever-
1
,
VOL. 12 No. 1
MANKIND SlGNlFlER
metasignifier ev i rPhase 1
,-hanu I
/.I
decay
I
Phase 2
,phan
-- -
I
I
decay
pi
-
-- --
-
c-
hanu
hanu
morphological
grasshopper species
crickets and grasshoppers
T,
~
various -other
5-r.
perceived similarity
mole c r i c k e t 1
various
‘--
SIGNIFIED
size and morphology
*insects*
-”.
insectsn
Phase 3 ~
inconsequential as food, collected by women and children
7
most frog species
I
FIG.1 Part of an hypothesized history of taxonomic change for the domain of ‘insects’ and frogs for Sane speaking people of Papua New Guinea. The rank of the taxon hartu has been progressively increased through time. This shift in rank has been contingent upon decay in the referents of hanu from connotative to denotative.
decaying hierarchy of ideological fields where increasing internal consistency is matched by increasing contradiction between those fields. This process is summarized in FIG. 2. I now want to map an interpretation of the evolution of zoological taxonomic systems onto this general model. I envisage three primary areas of ideological interest for animals, labelling these, in summary form, as objective, economic and (sensu luto) ritualistic. The ‘objective’ area includes that set of complementary metasignifiers dealing in similarities and contiguities of animal morphology, behaviour and ecology. The ‘economic’ area concerns metasignifiers dealing directly in energetic matters; in, for example, linked perceptions concerning the status of animals as food and clothing, in hunting or, perhaps, as hazards. The ‘ritualistic’ area covers metasignifiers whose complementarity springs from their concern with social existence, whether sacred or secular. Each area draws upon the same perceptual field. I am in difficulties immediately. In establishing these areas I have assumed a history of metasignification in which earlier more
diffuse areas have coalesced. The morphological or ecological attributes of animals and their status in myth or magic could act as separate areas of metasignification and give rise to distinct classificatory arrangements. There is, therefore, a degree of artificiality in my classification. Patterns of complementarity and contradiction for subsets of metasignifiers could be identified within each area and we could argue that a different permutation of sub-sets provides a more rewarding baseline. (Or even that I should ordinate rather than classify these sub-sets! ) I will let Australians defend my choice. For people from Groote Eylandt and Cape York there is evidence that separate classificatory fields occur; they have intersecting shallow taxonomies that are founded respectively in morphological and dietary criteria or, for plants at least, in patterns of ecological association (Chase and von Sturmer, in press; Hiatt, 1969; Thomson, 1939, 1946; Worsley, 1967). There is no good evidence that these classifications overlap comfortably with that of totemic labelling. For Australians, Worsley’s advice was that we ‘disconnect’ analyses of classificatory
VOL. 12 No. 1
MANKIND (A)
1
-
I
creation of set of complementary metasignifiers (an ideological f i e l d ) ?
I
perceptual f i e l d ( = e x t a n t socio-ecological environment)
of ideological fields a t t i m e 2 increasing consistency w i t h i n and contrast b e t w e e n f i e l d s of ideological fields at t i m e 1
perceptual f i e l d
FIG.2
(A) An hierarchical set of complementary metasignifiers, here defined as an ideological field, depicting relations between different and non-conflicting signified components of the extant socio-ecological environment. ( B ) Several hierarchical sets of complementary metasignifiers depicted by unbroken, broken and dotted lines respectively may yield a set of ideological fields (e.g. at time 1 ) . In turn, this may condition further metasignification within and between ,fields to generate, at time 2, a new set of ideological fields. This latter set will not emerge simply through consolidation within the previous set, it will entail disintegration of the old and a resorting of signifiers.
fields. It is a pity that this advice has often been taken as a general methodological rule.4 In my own exploration of Rofaifo animal classification I have found an elaborate taxonomy in which, with few excepti,ons, primary taxa may be characterized simultaneously in terms of the objective resemblance, dietary status and broad significance to social existence of included members. I have provided a detailed account elsewhere and have concluded, for Rofaifo, that taxonomic elaboration entailed progressive re-sorting of the three areas of ideological interest mentioned to produce an internally consistent zoology (Dwyer, n.d.). With one important exception re-sorting is simply the
consequence of articulating and incorporating similarities, detected between components from different ideological areas, into an emerging taxonomic structure. The exception concerns splitting of low ranking faunal categories which, in part, may be contingent upon expansion of segments of the perceptual field. Rofaifo taxonomy has a strong objective content. It looks good. It would appeal to scientific zoologists for it conforms well to taxonomic models derived from their practices. Taxonomies described for some other Papua New Guinean societies are messy by comparison. I refer particularly to those of Baktaman, described by Barth, and Kalam, described by Bulmer and co-workers (Barth,
E 161
VOL. 12 No. 1 1975; Bulmer, 1967; Bulmer and Menzies, 1972-3; Bulmer and Tyler, 1968; Majnep and Bulmer, 1977). Here, the taxonomic structure and, indeed, the taxa themselves may lack the depth or formal rigour apparent for Rofaifo. More significantly, objective, economic and ritualistic areas of interest exhibit different degrees of coalescence. Thus Barth explores and rejects the notion that patterns of metasignification concerning Baktaman social existence may be allied to classifications founded in objective criteria and his data show some separation of objective and economic foci. Bulmer, by contrast, identifies a synergistic relation between objective and dietary concerns in the patterning of Kalam taxonomy and argues that the latter imposes constraints upon the choice of metaphors that concern social existence. But, for Kalam, higher taxa do not coincide neatly with areas of ‘ritual’ signification and, on this count, taxonomic elaboration has not attained the level of ideological consistency that I found for Rof aifo. The evolutionary trajectory envisaged is thus one of progressive abstraction from a perceptual field initially apprehended as concrete. It proceeds through the transfer of meaning to ever-changing sets of complementary metasignifiers that, themselves, are patterned by contrast. If, as analysts, our perspective is respectively cultural or social then we may separate ideological or institutional configurations from the flow of socio-cultural activity. In the area of taxonomy I interpret the systems represented by Australians, Baktaman, Kalam and Rofaifo as different expressions of a programme through which changing patterns of metasignification progressively objectify nature. An endpoint for this programme would see the emergence of scientific taxonomic practice as ideology (or institution). The taxonomic systems mentioned remain distinctive configurations of metasignification-they are not moments of a frozen phylogeny. Indeed, my interpretation requires recognition that different taxonomic systems express meaning in incommensurate ways; that they articulate different perspectives on nature. When, as idiosyncratic ethnoscientists, we abstract from a culture some cluster of metasignifiers whose overt concern is with objective relations, and
MANKIND label this as the taxonomy, we lose the opportunity of arriving at useful crosscultural comparisons. The taxonomies we elicit are undoubtedly prescientific. But they do not grow through accretion into scientific form. Rather, they pass through layered sieves of metasignificatory possibility, disintegrating in the process of their recreation. The process (to pay homage to the French) is Marcel Marceau miming ‘masks’. In the preceding I commenced with three areas of ideological interest in animals. My interpretation for Rofaifo taxonomy required that the objective area took precedence as evolution proceeded; that economic and ritualistic areas provided sources for metasignificatory extension within an objective domain. This is evidenced by the fact that striking anomalies are few, are ones of inclusion and may be taxonomically rationalized in terms of economic or ritualistic criteria.5 Thus Rofaifo group eels and cassowaries with large mammals and link ‘insects’ and frogs but these apparently strange arrangements may be appreciated as expressions of high and low food values respectively. At higher levels of Rofaifo taxonomy emphasis upon objective criteria is weakened. Tambiah has described a different taxonomic system for people from Thailand (1969). It appears at least as elaborate in form and hierarchical depth as the Rofaifo system. Like the latter it sometimes confounds scientific objectivity in, for example, separating geese and wild ducks from other birds and allying them, respectively, with buffalo and elephant or in placing oysters, frogs and field crabs in a single category. Here economic considerations dominate the arrangement of taxa into more inclusive categories. Many recognized species are grouped according to domestication, their presence in field or water and, hence, their status as potential sources of everyday protein or their presence in forest. But, unlike Rofaifo, striking anomalies are relatively common and are ones of exclusion. Thus animals transgressing boundaries of economic domains may remain as unaffiliated taxa irrespective of apparent objective affinities; examples include lizards and rats which are found in houses although they are not domesticated or vultures and crows
MANKIND which, in almost indecent ways, may descend from the sky, where birds properly belong, to feed on carrion or alight on the roof of a house. Clearly, for these people the economic area has taken precedence in taxonomic evolution. The product, the extant taxonomy, is light years removed from that of Rofaifo though, for both societies, the process of emergence is founded in the one principle of change.
Contexts for taxonomic change Thus far, my story floats. I have designed feathers without flight, fur without bodytemperature regulation or evolved airbreathing quadrupeds and neglected to place them on land. There is no context for this programme of change. It has happened in vacuo; the taxonomies are adrift and functionless. Here we confront a paradox of evolution. We do not need and cannot have change of environment as either a necessary or sufficient cause of cultural change. The configurations of acts that we call human societies are environmentally constrained, not determined. The emergent sets of symbols that express meaning and which we call cultures have greater freedom. Yet culture, society and nature interact. Change in the last necessitates change in the othersadaptation is seldom avoidable-but change in the first, that is in culture, is not conditional upon change in the natural environment. How then may we situate an identified programme of taxonomic change contextually; how may we position it environmentally without specifying the outcome? The solution to this difficulty is, I think, inherent in the principle of change employed above. We may rephrase the question by asking how cultural change might specify environment. And my answer, I suspect, would have appealed to another Frenchman-Jean-Baptiste Lamarck (see Burkhardt, 1977: 144-51)! In establishing a relation between signified components metasignification both enhances contrast between those components and shifts the referents of the signifiers concerned. Hence metasignification modifies perception of the familiar. It redefines the familiar and may promote either expansion or contraction of perceptual fields. It does this irrespective of change in the phenomenal world. The en-
VOL. 1 2 No. 1 vironment as given need not change one iota yet each act of metasignification alters it for the people concerned and establishes a new milieu for the perception of similarities. The familiar itself is thus subject to abstraction and, for taxonomic change, there are two domains where the consequences of this will be crucial. They are human ecology, in the sense that ‘cultural anthropologists’ intend (e.g., Vayda and Rappaport, 1968) and ‘social structure’, for the former will constrain the content of taxonomic systems while the latter offers models for their structure. These are vast topics to burden this discussion with. They too must pass through layered sieves of metasignificatory possibilities to condense as ideologies and institutions. They too will coalesce from more diffuse, indeed more concrete spheres, as cultural meaning is articulated and extended through metasignification. Commencing as inseparable patterns of relations they may, in secure environments, be re-sorted, reified and shifted inevitably towards contrasting configurations of imaginary entities. We shall have reciprocity where once we had dynamic complementarity and have economics and nation states where once we had reciprocity. In the area of human ecology the central theme of change may be characterized as ‘internalization’. Increasingly, cultural change dictates the selection of relevant resources, drawing them into the ambit of culture itself to encourage a perception of nature as separate from culture. Human ecology shifts from crude dependence upon nature through diverse manipulative styles towards industrialization. It shifts from ecosystemic to analytic and, in concert, the perception of nature shifts from one of patterned relations to one of infinite commodity. It is this programme that constrains the content of taxonomic systems in demanding that ecosystemic survival strategies be wedded to a ‘science of the concrete’ while affording analytic strategies the luxury of abstraction. The nomenclaturally rich and hierarchically shallow taxonomies of gatherers and hunters store messages from a concrete world; those of other peoples may disavow such messages in proportion to their retreat from the phenomenal.”
VOL. 1 2 NO. 1 In the area of social structure I characterize the relevant theme of change as ‘formalization’. This is manifest as intragroup (or level) cohesion and intergroup (or level) discreteness increases in both segmentary and stratified arrangements of people. Social structure condenses as it shifts from acephalous groups through diverse segmentary patterns towards political centralization or from an absence of explicit forms of social control (i.e. dynamic complementarity) through modes of control patterned by and consequent upon psychological concomitants of reciprocity towards realization of an abstract authority imposed from above. But neither measures of social differentiation nor indices to complexity will sufficiently express an evolutionary theme of formalization (cf. Lomax and Berkowitz, 1972). This theme is not explicated simply as particular configurations of people; it is explicated through changing and emerging conceptions drawn from those configurations. It is the codification of social arrangements through a hierarchy of more or less clear-cut concepts that offers a model for the elaboration of rank within a taxonomy. Concepts of lineage, clan, phratry, tribe or kingdom or ones denoting strata within assemblages of people may be extended to express hierarchical rank within a taxonomy of animals. Having originated as expressions of perceived relations within or between groups of people they may now act as metasignifiers for groups of faunal categories. Where appropriate concepts are absent or ill-defined, so too elaboration of taxonomic structure will be inhibited.7 The taxonomic systems of the Papua New Guinea societies considered above are consistent with these suggestions. Baktaman live in the sparsely populated Western Highland fringes, they are the most dependent upon wild animals as a food source and have the least elaborate social structure. Their zoological taxonomy is shallow: its emphasis is nomenclatural and specific. Rofaifo are proficient pig husbanders of the densely populated Eastern Highlands and are the least dependent upon wild animals as food. They are a clan of several hundred people within a relatively complex segmentary arrangement of lineages, clans, phratries and tribe. Their zoological taxonomy is hierarchically rich and, interestingly, at the level
MANKIND of terminal taxa it frequently splits the given categories (i.e. species) of nature. For these societies, increased abstraction within the domain of taxonomy correlates with internalization in the area of human ecology and formalization in the area of social structure. The explosion of hierarchial depth enoountered in scientific taxonomy has similarly been founded in analogies with social structure. Linnaeus acknowledged this in establishing the ranks of class, order, genus, species and variety, and Lamarck despaired of it in commenting that ‘botanists had wanted to force nature “to arrange her productions as a general arranges his army: by brigades, by regiments, by batallions, by companies, etc.” ’ (Burkhardt, 1977 :54). Indeed, at the level of species, scientists have long since abandoned the concrete in, at one time, declaring species immutable types and, more recently, rationalizing them in evolutionary terms by obliging them to conform to the so-called ‘biological species concept’. Human ecology and social structure provide the context in which taxonomies evolve. They act, respectively, to constrain the content and to condition the structure of taxonomic systems. And they act in concert, within boundary conditions set by nature, to facilitate abstraction. For, in a world perceived as secure, there can be no escape from abstraction; in as much as one domain of human experience is patterned in abstraction, so that domain, by acting as a source for metasignification, nourishes abstraction elsewhere. This is the paradigm for change in the expression of cultural organization. It asserts that taxonomic systems shall store messages about organisms that condition perceptions of them and reflect a history of past relations. The storage system is lexical, it is patterned by contrast, but in its origins and its impact it is analogic. A knowledge of the (digital) rules implicit in the taxonomy used by another culture does not inform us how the people concerned will interact with the components of nature embraced by that taxonomy (cf. Vayda and Rappaport, 1968). The rules are artifactual; they comprise an ordering device that is contingent upon the progressive decay of past modes of metasignification. But an appreciation of taxonomic
MANKIND systems in an evolutionary frame can tell us something about ways in which the people concerned have, in the past, organized their relations with nature. And, as I argue below, it can help us understand why the styles of, and contexts for, animal metaphorization differ greatly between societies. Totemic labelling In the preceding account of taxonomic evolution there was an implication that components perceived as similar were often homologous or isomorphic (FIG. 1) . When faunal categories of low taxonomic rank are subsumed by higher order taxa, and where the latter may be specified by a small number of distinctive features, this implication may hold (cf. Berlin, 1976; Brown, 1977). But it is not invariant. If, as suggested above, concepts pertaining to social structure are used as models for conceptualizing rank (cf. genus, species, etc.) in taxonomy, then the process of metasignification entails perception of similarity across segments of domains that differ in degree of familiarity. For example, if the properties of discreteness and temporal continuity thought to characterize lineages are extrapolated to the faunal domain, then the familiar concept ‘lineage’, drawn from social structure, might be used to crystallize an as yet unfamiliar concept ‘species’ in taxonomy. Thus the essentially denotative function served by a signifier in one domain may promote emergence of a comparable connotative function in another. Metasignification will, therefore, often entail use of the more familiar to clarify or extend the meaning of the less familiar. This is what happens when the referents of animal categories are extended in metaphoric ways to explicate social meaning. Here the faunal domain represents the familiar or known while the domain of social existence is unfamiliar or mysterious. Animal metaphors enlarge social meaning, they order the mysterious; they need not simultaneously (though they may eventually) systematize the faunal domain that is their source (Barth, 1975:210-11; see also Beck and discussants, 1978). This account suggests alteration of my primary notion for change. FIG. 3 shows a modification appropriate to emergence of a general concept of species and generalizes
VOL. 1 2 No. 1 this to the circumstance of animal metaphorization. The essence of the latter is that relationships within the domain of social existence may be established through analogizing components of this domain with ones of the faunal realm and may be articulated by extending the referents of an appropriate faunal signifier. Where signifiers concerning components of the social domain are covert (e.g. non-lexical) or even absent animal metaphorization affords a powerful tool for the construction of social meaning. The model may be adapted to refer to ‘single’ components of social existence. The above may be illustrated by reference to totemic labelling; that is, the widespread practice among preliterate peoples of using the names of animal species (or other natural categories) to refer to classes of people. Here, statements of the sort ‘I am a cockatoo’, meaning ‘I, of this (class of people), am to all people as a cockatoo is to (a class of nature)’, lend clarity and meaning to perceived configurations of people by analogizing the latter to the familiar property of discreteness that characterizes the natural world, a property that is itself signified by naming and not by the names used. In an important sense therefore the particular label may be unimportant here. That label connotes a property common to all members of the domain from which it was drawn and it is this property that is now apprehended and signified in the domain of social existence. The fact that, in Australia, totemic labels may be selected, almost indiscriminately, from animate and inanimate domains (e.g. animal and plant species, heavenly bodies, stones, geographic features, the wind, diarrhoea and vomit) (e.g., Worsley, 1967), is not a matter for concern since the relevant properties may cross such boundaries. Indeed, the anthropological anxiety that totemic labelling is somehow less orderly because it transgresses ‘natural boundaries’ (i.e., those of twentieth-century scientists) would probably not have worried naturalists of the eighteenth century who divided ‘nature’s productions’ into three kingdoms-animal, plant and mineral-and recognized ‘species’ in each. There seems no reason to think that it worries the Australians. Totemic labelling is seldom as precious as the foregoing implies.8 The particular
MANKIND
VOL. 12 No. 1 connotative
denotative
increasing familiarity
*
