Semiotica 210 (2016), 215-234.
In this paper, the basic concepts of Peircean semiotics are derived from visual experience by the process of conceptual embodiment. We begin with embodiment of the universal Categories of Being that are accessible to thought or the universal Categories of Thought, which Charles S. Peirce defined and termed Firstness, Secondness and Thirdness. On this basis, we demonstrate conceptual embodiments of the Peircean typologies of dyadic relations, triadic relations and representations. The phenomenology of visual perception is modeled as a triadic typology of embodied mental processes which we term detection, localization and identification. Based upon the pioneering work of Lakoff and Johnson, we examine the role of visual embodiment in concept formation as inferred from linguistic expressions. We conclude that certain fundamental physical and relational concepts may be regarded as the embodied interpretants of visual signs.
Keywords: Peircean categories; visual signs; conceptual embodiment; linguistic signs; figurative signs
In previous studies, this author has presented a conceptual model of visual perception that is based upon semiotic principles first articulated by Charles Sanders Peirce (Cantor 2014a; Cantor 2014b). These principles include a triadic typology of ontological concepts or Categories of Thought which induce triadic typologies of binary relations and representations (“On a new list of categories,” EP1: 10). In its least general form, the Peircean concept of 'sign' may be thought of as a triadic relation involving a representamen or subject of the sign that refers to the object of the sign in a way that is specified by an interpretant, which expresses the meaning of the sign. In contrast, Lakoff & Johnson (1980, 1999) have developed a general model of the mental processes involved in concept formation, based upon research in cognitive science and cognitive linguistics. In this model, conceptualization involves the unconscious formation of schematic neural representations as a consequence of sensorimotor experience. The succession of concepts in the process of thought is determined by unconscious modes of inference, which are also products of sensorimotor experience. In this study, the term conceptual embodiment will refer to the unconscious mental processes in which concepts or implications are formed as a result of perceptual or motor activity. Lakoff & Johnson (1980, 1999) have demonstrated that the existence of such unconscious mental processes may be inferred from their expressions in language. Such expressions are said to be literal if they are ‘direct’ representations of sensorimotor experience and figurative if they are ‘indirect” representations, in the sense of having been derived from literal expressions. In this study, we present a semiotic model of visual perception in which the formation of interpretants, i.e. the meaning of visual signs, is a product of conceptual embodiment. We shall assume that most intentional or non-reflexive motor activity is under visual control, i.e. visual guidance and feedback, while habitual motor activity is learned under visual guidance. We shall demonstrate that certain abstract physical and relational concepts may be regarded as embodied interpretants of visual signs. We begin by defining embodiments of the Peircean ontological categories from which are derived triadic typologies of relations and signs.
In his seminal paper of 1867 (“On a new list of categories, EP1: 10”), Charles S. Peirce identified, by phaneroscopic analysis, three irreducible modalities of Being or Thought. For maximum generality of expression, Peirce later termed these categories Firstness, Secondness and Thirdness (see Ransdell [Arisbe] for an overview of Peirce’s changing terminology for the Categories of Being or Thought). From a human experiential perspective, Firstness may be regarded as the attribution of a property to an entity, Secondness as an opposition between one entity and another, and Thirdness as a mediation between two entities that are in opposition. Hence, 1st Category entities involve a mental process of ‘attribution,’ 2nd Category entities a mental process of ‘opposition,’ and 3rd Category entities a mental process of ‘mediation.’ We note that mediation between two entities presupposes an opposition and an opposition between entities presupposes attributions that distinguish between the two. This sequence of conceptual entailments exemplifies the categorical inclusion rule as defined by Liszka (1996: 46), after Peirce. We propose that Peirce abstracted his universal Categories of Thought from visual experience. In section 5, we demonstrate basic mental operations in the phenomenology of perception that may be regarded as embodiments of the Peircean Categories of Thought.
In his paper, “On a new list of categories,” Peirce also defined three types of binary relations between successive thoughts. In the context of visual perception, we define a binary relation to be the perception of two distinct relates in the same moment of awareness. It follows that a binary opposition is a binary relation with one relate that is marked for selective attention while the other relate is unmarked (cf. Cantor 2012, on markedness as a sign of difference). The Peircean categorical relations were derived from the Peircean universal Categories of Thought and were initially characterized as relations of ‘similarity,’ ‘contiguity’ and ‘convention,’ ‘custom’ or ‘habit.’ However, aspects of this typology of relations had been identified by philosophers before Peirce, the most influential being Aristotle and Hume. Aristotle (1941: 612, 614) was possibly the first to observe that the process of recollection is governed by relations of ‘similarity,’ ‘contrariety,’ ‘contiguity’ or ‘custom’ between successive thoughts. Hume (2002: 13,416) inferred that there are only three fundamental principles of association between successive thoughts, which he termed ‘resemblance,’ ‘contiguity’ and ‘causation.’ Since ‘contrariety’ has the logical connotation of ‘contrast’ and ‘contiguity’ has the spatial connotations of ‘contact’ or ‘proximity,’ both ‘contrariety’ and ‘contiguity’ presuppose the concept of ‘opposition.’ We recall that Hume (2002: 413) regarded causation as a type of ‘custom’ or ‘habit’ that depends upon the recollection of the ‘constant conjunction’ and ‘regular succession’ of cause and effect. We note that the concept of ‘custom,’ as applied to a population, is a social convention and that the concept of ‘habit,’ as applied to an individual, is a disposition that is acquired by repeated experience. Hence, both ‘custom’ and ‘habit’ presuppose an agreement produced by repetition, which we shall term ‘conventionality.’ Furthermore, Peirce acknowledged the influence of Hegel in his formulation of a triadic typology of Categories of Thought, which parallels Hegel’s “three stages of thought”: Thesis, Antithesis and Synthesis (“Pragmatism,” EP 2: 428). Based upon these considerations, we propose a slight modification of the original Peircean typology of binary relations:
- an opposition between common attributes of its relates, e.g. color, texture or contour, which may be termed ‘contrast.’
- an opposition between common places of its relates, e.g. ‘contact’ or ‘proximity,’ which may be termed ‘contiguity,’ or
- an opposition between common states of its relates, e.g. ‘motion’ or ‘function,’ which may be termed ‘reaction
This generalized typology of binary relations, which includes the typologies of Aristotle, Hume and Peirce, may be summarized as follows:
According to Peirce, the process of signification is determined by a triadic relation. Numerous general definitions of the ‘sign’ concept may be found in his writings (Deely 2014). However, in visual semiotics it is sufficient to use a specific concept of ‘sign’ that is based upon Peircean principals, which we shall term a ‘mental sign.’ We define a mental sign to be a triadic relation in which a first relate (the representamen) is marked as a substitute for a second relate (the object) in a given context while a third relate (the interpretant) mediates or conceptually connects the other two. Furthermore, in a mental sign, the mind of the interpreter is a ground for all of its relates. In this way, a visual sign may be thought of as a triadic relation consisting of a binary opposition between a perception and a recollection that is mediated by an interpretation (cf. Cantor 2011). In visual semiotics, the interpretant of a sign is a linguistic expression of its meaning. Hence, the Peircean typology of binary relations (section 3) induces a typology of triadic signs such that:
The perception of visual phenomena begins with the subjective partition or segmentation of the visual field into discrete colored regions (cf. Cantor 2014, after Palmer 1999). The interface between contiguous regions is perceived as color contrast. Visual figures are perceived as colored regions that have edges, in the same way that we expect physical objects to have palpable surfaces. The contours of a visual figure are subjective attributes of its edges. Hence, the colors and contours of a visual figure are not intrinsic properties of the object it represents, but depend upon ambient lighting, visual physiology and optic perspective of the viewer. The edges of a visual figure are completely surrounded by regions of contrasting color that collectively constitute a figural background or ground. Hence, a visual figure presents a triadic relation to the mind of the viewer. This relation consists of a binary opposition between a figural ‘inside’ and a figural ‘outside’ that is mediated by a figural ‘edge.’ This relation is irreducible in the sense that the meaning of each relate depends upon the meanings of the other two, e.g. we cannot conceive of an inside without accompanying concepts of an outside and a mediating edge, etc. It is likely that this fundamental triadic configuration constitutes the visually embodied prototype for the abstract concept of an irreducible triadic relation.
5.1 Embodied detection
In the phenomenology of visual perception, the process of detection entails the formation of visual figures with the attribution of visual qualities by the viewer. In particular, visual detection entails the attribution of edge contrast and edge continuity that define the spatial extension and unitary nature of a visual figure, which are characteristics of a physical object. Such attributions of figural properties make possible subsequent processes of localization and identification. Hence, visual detection may be regarded as a 1st Category perceptual process.
5.2 Embodied localization
The process of visual localization is governed by the nature of light and the situation of the viewer, i.e., that light moves in straight lines and that the line of sight of a visual object is determined by the orientation of the body, head and eyes of the viewer relative to the object. Consequently, opaque objects that are closest to the viewer in a line of sight will prevent the formation of visual figures of more distant objects. In general, visual localization entails an opposition of place between the viewer and a visualized object, where to be in spatial opposition means to be located at opposite ends of a line of sight. Hence, visual localization is a 2nd Category perceptual process.
5.2.1 Egocentric localization
The embodied Self is at the center of the perceptual world of an individual. An individual mind partitions its perceptual world into an embodied Self and what is regarded as non-Self. The primary process of localization of non-Self objects employs an egocentric spatial reference system (cf. Lakoff & Johnson 1999: 30-34). The concepts of ‘position’ and ‘orientation’ are derived from visual experience and constitute the basic parameters of this reference system. The position or placement of an object may be ‘inside’ or ‘outside’ the embodied Self. If it is ‘outside,’ it may be ‘on’ or ‘off’ the surface of the body. If it is not ‘on’ the surface, it may be ‘near’ to or ‘far’ from the embodied Self. The orientation of an ‘outside’ object is conventionally defined by its opposition to a named surface of the embodied Self. Hence, in the erect posture, the forehead (and eyes) are directed toward the ‘front,’ the top of the head is directed ‘above’ and the soles of the feet ‘below.’ Canonical aspects of the body are defined in opponent pairs, e.g. ‘front’-‘back,’ ‘above’-‘below,’ ‘right’-‘left’. While the orientation of an ‘outside’ object is directly perceived in the layout of a visual scene, its position in depth is inferred by the interpretation of depth signs (Cantor 2014a).
5.2.2 Allocentric localization
A secondary (indirect) process of object localization employs an allocentric spatial reference system, i.e. localization relative to an entity other than the embodied Self. An allocentric reference system is a mental projection of an egocentric system onto a non-Self object (cf. Lakoff & Johnson 1999: 34-35). Hence, we conceive of visual figures as representing objects with surfaces that separate insides from outsides in the same way that we experience an opposition between the ‘inside’ and ‘outside’ of the embodied Self. The aspect of objects with which we usually interact is called the ‘front.’ For unfamiliar stationary objects, the aspect in the line of sight is also called the ‘front.’ For unfamiliar moving objects, the ‘front’ is regarded as the direction of motion. The ‘up’-‘down’ directions are defined relative to the ground that supports the embodied Self.
5.3 Embodied identification
Identity is an attribute of individuals or categories that distinguishes them from all others in a given context. Designation by naming facilitates the process of identification. Both the name of an individual and the appellation of a category are conventional representations or symbolic signs (cf. Jakobson 1990: 318-319). Two temporal modes of identity have been distinguished in contemporary ontological studies (cf. Cantor 2010, for semiotic applications and references):
On this basis, a visual entity may be designated as a continuant or an occurrent. Visual continuants are entities that have spatial boundaries, i.e. objects or substances. Their identity is assumed to be invariant under entropic change, e.g. material wear or biological aging. In contrast, visual occurrents are entities that have temporal boundaries, e.g. events or processes. Both continuants and occurrents may be conventionally divided into parts, in the same way that we distinguish parts of the embodied Self. In visual perception, identification of an object involves the recognition of a conventional partition, where the parts, which have conventional or fiat boundaries, are perceived as constituting a unitary whole. We also note that visual identification of an object entails its localization within a spatial context and the detection of its visual qualities. This is another manifestation of the categorical inclusion rule (section 2). Hence, visual identification is a 3rd Category perceptual process.
We assume that the existence of a human mind entails the existence of a living body, i.e. that minds are embodied. Hence, conceptual meaning is dependent upon conceptual embodiment. This implies that the interpretants of visual signs must depend upon the visual memory of the viewer. In this section, we propose a triadic typology of visually embodied concepts which is derived from the Peircean Categories of Thought. This typology is basically a reinterpretation of the fundamental work of Lakoff & Johnson (1980, 1999) on the formation and expression of spatially grounded concepts. We assume that embodied concepts may be expressed either literally or figuratively. A literal linguistic expression is a phonic representation of an unconscious image schema formed by direct sensorimotor experience and subjectively experienced as a literal concept. Hence, a literal concept is directly derived from sensorimotor experience. A figurative linguistic expression is a phonic representation of an image schema that is unconsciously associated with a literal image schema and subjectively experienced as a figurative concept. Hence, a figurative concept is indirectly derived from sensorimotor experience.
6.1 Literal embodiment
Literally embodied spatial concepts are products of visual perception. Categorical modes of visual embodiment may be characterized as ontological, topological or morphological embodiments.
6.1.1 Ontological embodiment
Ontological embodiment entails the attribution of visual properties such as extension, color and shape to a visual object (section 5.1). The edges of a visual figure trace a simple closed curve that partitions the visual field into an ‘inside’ and an ‘outside.’ A visual figure is assumed to represent a physical object that exists outside of the embodied Self, i.e. an entity that has the same ontological properties of extension and substance as the embodied Self. Since ontological embodiment involves the attribution of visual properties, it is a 1st Category mode of visual embodiment.
6.1.2 Topological embodiment
Topological embodiment entails the perception of a spatial opposition between the embodied Self and a visual object (section 5.2.1). By a process of mental projection, a visual object may be regarded as a substitute for the embodied Self in the formation of spatial oppositions (section 5.2.2). Spatial oppositions may be perceived as static or dynamic relations. Since topological embodiment involves the perception of a spatial opposition, it is a 2nd Category mode of visual embodiment.
6.1.3 Morphological embodiment
Morphological embodiment may involve either the exterior or the interior of a visual object. Exterior morphological embodiment is derived from the conventional partition of the embodied Self, e.g. a head, neck, body, etc. Hence, by mental projection, visualized objects are conventionally divided into parts, where contiguous parts share a conventional boundary and constitute a unitary whole (Lakoff & Johnson 1980: 58). Interior morphological embodiment involves the concept of containment. A ‘container’ may be conceptualized as an object that permits the movement of another entity (a substance or an object) along a path connecting its ‘outside’ with its ‘inside’ (cf. Lakoff & Johnson 1999: 30-34, on spatial relation concepts). It is likely that the embodied Self, with its functions of ingestion and excretion, serves as the prototype for the literal concept of ‘container’ (cf. Lakoff & Johnson 1980: 29). Since both exterior and interior modes of morphological embodiment involve the mediation between conventional parts or conventional aspects of objects, they are both 3rd Category modes of visual embodiment.
6.2 Figurative embodiment
In the model of conceptual embodiment proposed by Lakoff & Johnson (1999: 77, 128), subjective and abstract concepts are derived from literal concepts by unconscious modes of inference. The meaning of a concept is regarded as the subjective experience that accompanies the activation of an image schema in the cognitive unconscious, where an image schema is a dynamic neurophysiological representation of sensorimotor experience, i.e., the activity pattern of a neuronal network that is formed as a result of perceptual experience. While literal concepts are direct expressions of sensorimotor experience, figurative concepts are indirect expressions that are derived from literal concepts. This implies that figurative linguistic signs are derived from literal linguistic signs. A literal linguistic sign consists of a binary opposition between a linguistic expression and a literal image schema, which is mediated by an interpretant. The literal image schema is formed in the cognitive unconscious by direct sensorimotor experience and the interpretant or meaning of the sign is expressed in language. A figurative linguistic sign consists of a binary opposition between a figurative linguistic expression and a figurative image schema, which is mediated by an interpretant. The figurative image schema is formed in the cognitive unconscious by Categorical association (section 3) with a literal image schema. The interpretant or meaning of the sign is expressed in figurative language (cf. Jakobson 1990: 417, for a dyadic concept of figurative signs). Hence, a figurative interpretant may be expressed as a relation of similarity, contiguity or conventionality. The interpretant of a figurative visual sign may be regarded as a figurative identity relation. In a literal identity relation, every property of one relate is shared with the other relate and vice versa (cf. Tarski 1994: 50, on identity according to Leibnitz). In visual semiotics, a relation of figurative identity may be defined by
In this way, the Peircean categorical typology of binary relations (section 3) induces a triadic typology of figurative linguistic signs. By conventional usage, figurative linguistic signs may be designated as metaphoric, metonymic or synecdochic signs (cf. Lakoff & Johnson 1999: Table 4.1, for examples).
6.2.1 Metaphoric signs
The object of a metaphoric sign is associated with the object of a literal sign by an unconscious relation of similarity. Hence, a metaphor may be regarded as an indirectly iconic linguistic sign.
6.2.2 Metonymic signs
The object of a metonymic sign is associated with the object of a literal sign by an unconscious relation of contiguity. Hence, a metonym may be regarded as an indirectly indexical linguistic sign.
6.2.3 Synecdochic signs
The object of a synecdochic sign is associated with the object of a literal sign by an unconscious relation of conventionality. Hence, a synecdoche may be regarded as an indirectly symbolic linguistic sign, e.g. a sign in which a part substitutes for a whole or vice versa.
In previous sections, we have defined triadic typologies for embodied linguistic expressions, both literal and figurative, in terms of the universal Categories of binary relations (section 3). In the sections that follow, we cite examples of conceptual embodiment, specifically for the abstract concepts of ‘space,’ ‘time’ and ‘causation.’
7.1 Embodied concepts of space
Embodied concepts of space originate in the visual relations of ‘opposition’ and ‘containment.’
7.1.1 Literally embodied concepts of space
Literal concepts of space entail visually perceived relations between the embodied Self and non-Self objects. These relations entail the opposition between the two ends of a line of sight. A typology of literally embodied spatial relations is given in Section 5.2. In contrast, Leibnitz formulated the concept of space as a complex of relations between coexisting objects (Jammer 1954: 48).
7.1.2 Figuratively embodied concepts of space
Figuratively embodied concepts of space entail a visual relation of ‘containment.’ The place of an object is conceived as containing the object and the object is conceived as contained in its place. Abstract space may be locally conceptualized as a place that contains an object or globally conceptualized as the aggregate of all potentially occupiable places (cf. Jammer 1954: 15, after Aristotle). Hence, a ‘place’ may be conceptualized as a 'part' of space. In the same way that the movement of visual figures is assumed to be relative to a stationary ground (Palmer 1999: 504), an abstract concept of space may be figuratively conceptualized as a stationary container that admits movement of the objects within it. This global concept of space may be regarded as the embodied prototype of the intuitive Newtonian concept of Absolute Space (cf. Jammer 1954: chapter 3, on the historical development of the concept of Absolute Space).
7.2 Embodied concepts of time
Concepts of time are derived from the visual perception of events, i.e. time is conceptualized as a change of place or change or appearance. What follows is an adaptation of the work of Lakoff & Johnson (1999) on embodied concepts of time.
7.2.1 Literal concepts of time
Literal concepts of time are derived from the visual perception of regularly repeated events. Such events may be natural (phases of the Moon) or artifactual (readings of a clock). Literal concepts of time such as ‘moment’ and ‘duration’ are expressed by the indexing or counting of events.
7.2.2 Figurative concepts of time
Most figurative concepts of time are derived from the visual perception of relative motion between an observer and a continuant, i.e. an object or fluid:
Hence, abstract time may be visually conceptualized as either a stationary or moving entity. We recall that, in a visual scene, a figure is perceived as either stationary or moving relative to its ground, while its ground is tacitly assumed to be stationary (Palmer 1999: 502). Hence, in Moving Time expressions, time is conceptualized as a moving figure and the stationary observer is part of its ground. In Stationary Time expressions, a moving observer is conceptualized as a figure with time as its stationary ground. Hence, Moving Time and Stationary Time expressions involve dual concepts of time (Lakoff & Johnson 1999: 148-149). We have seen that the abstract concept of time is spatialized in the figurative expressions that represent time as a stationary or moving continuant. In Stationary Time expressions, the path of a moving observer may be thought of as an imaginary line in space. Therefore, the place of the moving observer at any instant corresponds to a ‘point’ on this imaginary line. Similarly, a duration of movement corresponds to an ‘interval’ on the imaginary line. Hence, imaginary spatial ‘points’ represent temporal ‘instants’ by metonomy and imaginary spatial ‘intervals’ represent ‘durations’ by metaphor. In Stationary Time expressions, the direction of movement represents the subjective ‘direction’ of time. At any instant, an egocentric present is represented by the ‘place’ of the moving observer, while the future is in ‘front’ and the past is in ‘back’ of the observer (cf. section 5.2.1). To paraphrase Lakoff & Johnson (1999: 146), these examples of figurative expressions demonstrate a ‘systematic polysemy between spatial and temporal meanings of the same words.’
7.3 Embodied concepts of causation
In this section, we present a model of embodied causation that is based upon the concept of causation as articulated by Hume (2002) and the concept of embodiment as developed by Lakoff & Johnson (1999). According to Hume (2002: 111-112; 413), causation is a relation of ‘constant conjunction’ and ‘regular succession’ between a cause and an effect. Hence, cause and effect exist in spatiotemporal proximity (a cause accompanies its effect) and in temporal succession (a cause precedes its effect). Furthermore, causation is a relation of necessity that is derived from custom or habit. It is embodied as an expectation that is produced by repetition. Lakoff & Johnson (1999: 177) have shown that different figurative concepts of causation may be derived from a literal prototype.
7.3.1 Literal causation
The literal prototype of embodied causation involves human agency. For example, manipulation may produce a visual change in an object. This may be a change of place or a change of appearance (cf. Lakoff & Johnson 1999: 177). This implies that a cause determines an effect in the same way that intentional activity has an expected outcome. This conceptual prototype of causation also entails the expectation of the physical ‘exertion’ or ‘effort’ that accompanies motor activity. Hence, the concept of ‘necessity’ is embodied as an expectation. We note that the root meaning of the word ‘exertion’ is ‘to thrust out’ and that the words ‘effort’ and ‘force’ have the same root meaning of ‘strong.’ Hence, embodied ‘necessity’ entails the expectation of an accompanying effort and this effort is reified as a ‘force.’ In this way, ‘force’ is conceptualized as an entity that determines a change of motion while motion is perceived as a change in place. In summary, the conceptual prototype of embodied causation involves a triadic relation consisting of a binary opposition between the embodied Self (an agent) and an object (a patient) that is mediated by a kinesthetic sensation (a force).
7.3.2 Figurative concepts of causation
Lakoff & Johnson (1999: 196) have demonstrated that two distinct concepts of causation are implied by the spatially grounded metaphors ‘states are locations’ and ‘attributes are possessions,’ where ‘states’ are conditions of being and ‘attributes’ are imputed properties. These metaphors have the following consequences:
Hence, both visually embodied and spatially grounded metaphors have parallel entailment patterns. Lakoff & Johnson (1999: 198-199) define figure-ground relations for figurative expressions by analogy with visual figure-ground relations. We recall that visual figure-ground segregation entails a binary opposition between a visual figure that is marked for selective attention and a visual background, which is unmarked (section 5). We tacitly assume that visual figures move in relation to a stationary background. As a consequence of the ‘state is location’ metaphor, a state may be regarded as a figure that moves in relation to its ground. This implies a concept of causation in which the affected entity is a figure and the effect is a changed ground. As a consequence of the ‘attributes are possessions’ metaphor, an attribute may be regarded as a figure that moves in relation to an object considered to be its ground. This implies a concept of causation in which the affected entity is a ground and the effect is a changed figure. Hence, based on their figure-ground structure, the visually embodied and spatially grounded metaphors ‘states are locations’ and ‘attributes are possessions’ entail dual concepts of embodied causation (Lakoff & Johnson 1999: 195, 198). In both concepts, a cause is expressed as a ‘force’ and an effect is expressed as a movement.
7.4 Non-visual embodiment
In this study, we are primarily concerned with the role of visual perception in the formation of concepts. However, many concepts are embodied with non-visual modes of perception. For example, subjective concepts are frequently embodied with sensations of physical contact produced by the manipulation of objects. Figurative concepts of ‘softness’ and ‘hardness’ are embodied as sensations of yielding and non-yielding to physical pressure. Concepts of ‘smoothness’ and ‘roughness’ are embodied by regular and irregular sensations of touch. Concepts of ‘sharpness’ and ‘dullness’ are embodied by opposite sensations of pain and pressure. Subjective concepts of ‘warmth’ and ‘coldness’ are embodied by the thermal sense. We have seen that the abstract concept of ‘force’ is embodied by the kinesthetic sense, etc.
We assume with Lakoff & Johnson (1999: chapter 2) that conceptualization originates in unconscious mental processes which are collectively referred to as the cognitive unconscious. A concept is then defined as the subjective experience that accompanies the activation of an image schema in the cognitive unconscious. In Lakoff & Johnson (1999), image schemas are defined as products of sensorimotor experience. Since intentional motor activity is guided primarily by visual perception, we regard an image schema as the activity pattern of a neuronal network or neural assembly the connectivity of which is formed by repeated or salient visual experience. Also, a conceptual field is defined as the aggregate of concepts that may be inferred from a particular image schema. It is likely that most cognitive processes are independent of conscious thought. In everyday experience, unconscious mental processes are spontaneously revealed to awareness and expressed in language (cf. Hadamard 1954). Thus, the cognitive unconscious may be regarded as a primary semiotic modeling system and language as a secondary semiotic modeling system (cf. Sebeok & Danesi 2000). It is likely that conceptual inference, which is consciously experienced as an association between antecedent and consequent thoughts, occurs in the cognitive unconscious. Inference in this sense is assumed to be due to an unconscious correspondence within or between conceptual fields. Charles S. Peirce conceived of conscious inference as a triadic relation in which an antecedent expression (a case) and a consequent expression (a result) are related by a rule (“Deduction, induction, hypothesis,” EP1: 188). He postulated three fundamental modes of inference, which he termed ‘abduction,’ ‘deduction’ and ‘induction.’ In section 3, we have seen that the Peircean Categorical typology of binary relations may be derived from three universal Categories of Thought which we have referred to as ‘attribution,’ ‘opposition’ and ‘mediation.’ In what follows, we demonstrate that a Categorical typology of embodied inference, i.e. embodied ‘abduction,’ ‘deduction’ and ‘induction,’ may be derived from the Peircean typology of binary relations. We begin with a characterization of informal inference that is based upon Peircean principles. Following this, we demonstrate likely embodiments of the informal modes of inference.
8.1 Informal inference
Informal modes of inference are used in everyday thought and are also the basis for scientific or formal inference (cf. Cantor 2011). In informal abduction, the consequent is experienced as a guess that has the subjective value of a possibility (“On the logic of drawing history from ancient documents,” EP2: 107 “The nature of meaning,” EP2: 216). In informal deduction, the consequent is experienced as an association that has the subjective value of a necessity (“The nature of meaning,” EP2: 216; “Pragmatism as the logic of abduction,” EP2: 233). In informal induction, the consequent is experienced as an expectation that has the subjective value of a probability (“The nature of meaning, ” EP2: 216: “Pragmatism as the logic of abduction,” EP2: 233).
8.2 Embodied inference
According to Lakoff & Johnson (1999: 102), embodied conceptualization involves a sequential activation of nonconscious and conscious neural processes. According to Peirce, the universal Categories of Thought determine three fundamental modes of inference. In what follows, we sketch a Categorical model of embodied modes of inference.
8.2.1 Embodied abduction
In embodied abduction, there is awareness of the consequent while the antecedent and the rules of inference remain in the cognitive unconscious. We suggest that the embodied rules of abductive inference involve the unconscious detection of concepts that belong to conceptual fields of both the antecedent and the consequent. In this case, abduction would involve the unconscious detection of a conceptual similarity (“On the logic of drawing history from ancient documents,” EP2: 106). According to Peirce, abduction is the only mode of inference that generates ‘new’ concepts (“The nature of meaning,” EP2: 216).
8.2.2 Embodied deduction
In embodied deduction, there is awareness of the antecedent and the consequent while the rules of inference remain in the cognitive unconscious. We suggest that the embodied rules of deduction involve the unconscious detection of the consequent within the conceptual field of the antecedent. Hence, embodied deduction is fundamentally the inference of a part from a whole (“On the logic of drawing history from ancient documents,” EP2: 96). Hence, Peirce claims that deductive inference adds nothing ‘new’ to what is already known (“Deduction, induction, hypothesis,” EP1: 199).
8.2.3 Embodied induction
In embodied induction, there is awareness of the antecedent and the consequent while the rules of inference remain in the cognitive unconscious. The antecedent consists of the recollection of the repeated occurrence of an event under certain conditions. The consequent is the expectation of the reoccurrence of the same event under the same conditions. The unconscious rule of inference is ‘the more frequently an event occurs, the more likely it will reoccur under the same conditions.’ This rule of inductive inference is probably embodied with the mental process of habit formation. Hence, embodied induction is fundamentally the inference of a general from a particular or the inference of a whole from a part (“Deduction, induction, hypothesis,” EP1: 189).
In this section, we regard a text as a visual representation of speech, where speech is a phonic representation of the subjective experience of thought (cf. Lakoff & Johnson 1999: 10-14). Hence, the meaning of a text is accessed by an unconscious process of “intersemiotic translation,” i.e. by the translation of visual signs into immediately interpretable phonic signs (cf. Jakobson 1987: 429). In the discussion that follows, we present a categorical typology of intersemiotic translations in which the visual form of written language embodies the meaning of spoken language.
9.1 First category intersemiotic translation
First category intersemiotic translation entails a similarity relation between perceived and recalled visual forms. Such a relation may be expressed as visual lengthening and interpreted as an ‘increase,’ in a general sense. Hence, plurality may be visually represented by the addition of a phoneme or suffix to a root noun, i.e. by elongation of a written form. In English, this involves the addition of ‘s’ or ‘es’ to the end of a written form (Jakobson 1990: 414). The gradation in meaning of the positive, comparative and superlative degrees of adjectives and adverbs is visually represented by a graded increased in the length of suffixes attached to a root form, e.g. long, longer, longest or fast, faster, fastest (Jakobson 1990: 414). Concepts of multiplication, continuation or intensification may be visually represented by the repetition of a syllable in a word or a word in a sentence (Lakoff & Johnson 1980: 157).
9.2 Second category intersemiotic translation
Second category intersemiotic translation entails a visual opposition between written forms which conveys the sense of ‘precedence’ or ‘dominance.’ In general, ‘the order of occurrence of physical events determines the order of representation of narrated events’ (after Peirce, as quoted in Jakobson 1990: 157). In a proposition, the subject indicates what the proposition is about and the predicate describes the subject (“The categories defended,” EP2: 172). In the same way, the localization of a figure in a visual scene precedes its identification (cf. section 5). In a declarative sentence, the subject acts while the object is acted upon (Jakobson 1990: 413). In conditional sentences, the conditional clause necessarily precedes the conclusion (Jakobson 1990: 413) in the same way that a cause necessarily precedes its effect. When ranked concepts are expressed in a sentence, the concept that is ranked as ‘first’ is placed before the concept that is ranked 'last' (Jakobson 1990: 412; Lakoff & Johnson 1980: 133). Within a sentence, when one concept has an effect on another, the strength of the effect is indicated by the distance between the corresponding words, i.e. when the words are close to one another, the effect is ‘strong’ and when the words are far apart, the effect is ‘weak’ (Lakoff & Johnson 1980: 128).
9.3 Third category intersemiotic translation
Third category intersemiotic translation entails conventional relations between written linguistic forms. A written text consists of a linear ordering of visual forms, i.e. syllables, words and sentences, that represents the temporal order of speech events. The spatial orientation of the lines of a text, the order in which the lines are read and the order in which the elements of each line are interpreted are all conventions maintained by a speech community.
We have described ways in which the basic concepts of Peircean semiotics may be derived from visual experience by the process of conceptual embodiment. In general, embodiment may be regarded as a mental process by which concepts are formed either directly or indirectly from memories of sensory or motor experience. We began with an examination of the metaphysical Categories of Being which Charles S. Peirce termed Firstness, Secondness and Thirdness. These abstract concepts are embodied in visual experience as the mental operations of Attribution, Opposition and Mediation. Based upon these universal Categories of Thought, we have proposed a slight modification of the Peircean triadic typology of binary relations, which we have characterized as Similarity, Opponency and Conventionality. We have suggested that the Peircean concept of ‘sign,’ as an irreducible triadic relation, is embodied in visual experience. Furthermore, the spatial opposition between the inside and the outside of a visual figure, which is mediated by a figural edge, may be conceptualized as the opposition between the embodied Self and the visual world, which is mediated by the sensory interface between the body and the visual world. We suggest that the perception of visual signs in visual experience is based upon the embodied mental processes of Detection, Localization and Identification. George Lakoff and Mark Johnson conclude from an analysis of linguistic expressions that conceptualization depends upon the embodiment of sensory and motor experience. In this study, we have emphasized the function of visual embodiment in concept formation. As examples, we cite visual embodiments of the abstract physical concepts of space, time and causation and the relational concepts of containment, implication and interpretation. We conclude that certain fundamental physical and relational concepts may be regarded as the embodied interpretants of visual signs.