Recent derivatives with the suffix -less: a change in progress within the category of english privative adjectives?
The aim of the present paper is to offer a cognitive grammar account of newly coined privative adjectives such as paperless (office) and cordless (telephone). (2) First, however, I will briefly characterize the overall network of constructional schemas for privative adjectives, concentrating on the semantics of prototypical patterns within it (section 2). (3) Referring to this network, I will argue that the new derivatives with the suffix -less impose a construal of the profiled conceptual content which substantially differs from that evoked by new derivatives with the suffix -free, such as acid-free (paper) or focus-free (camera) (section 3). On these grounds, I will suggest that the adjectives of the paperless and cordless type are likely to mark a change in progress within the category of privative adjectives as a whole (section 4).
2. The complex category of privative adjectives: An overview of main generalizations
From the perspective of cognitive grammar, privative adjectives have the status of a complex category, i.e. an assembly of interconnected symbolic structures. As a morphological category, the category of privative adjectives is represented in the form of a schematic network which subsumes not only specific expressions, but also several patterns, called "constructional (sub)schemas" (see Figure 1 in G6rska 1994: 415), all of which have the status of a "unit", i.e. a structure mastered by a speaker as a cognitive routine, so that "it can be employed in largely automatic fashion, without requiring attention to its individual parts or their arrangement" (Langacker 1991: 555).
Well-entrenched specific expressions in -less and -free, just like constructional (sub)schemas whose extraction (at different levels of abstraction) they have supported are the network's "nodes", which are interlinked via two main types of "categorizing relationships": elaboration/instantiation or else extension. In the case of semantic structures, (4) the former resides in establishing the relationship of schematicity between the relevant two senses: the two semantic structures are fully compatible in their specifications, differing only in the degree of precision and detail with which they are characterized. The categorizing relationship of extension, on the other hand, involves some conflict in the relevant specifications. Semantic extension is, moreover, "invariably based on some perception of similarity or association between the original (sanctioning) sense and its extended sense" (Langacker 1987: 157). (5) For example, the adjective childless instantiates/elaborates one of the most salient subschema of the -less subcategory, while the adjective sugar-free is an instantiation/elaboration of the most salient subschema of the -free subcategory. On the other hand, the nonce adjective child-free, as in:
(1) We are child-free for the weekend!
Would be categorized as an extension of the prototypical pattern for the adjectives in -free, since its semantic characterization has a lot in common with the specification of this schema (it designates a positively evaluated state which obtains due to some intentional actions of a human agent); at the same time, it departs from any schema with the suffix -free in that this suffix (unlike -less, cf. childless) is normally not integrated with nominal bases designating relatives.
It needs to be emphasized that the nodes and categorizing relationships in a network are not of equal status. They vary in terms of two parameters: cognitive salience or "entrenchment" and cognitive "distance". The latter parameter is measured in terms of the degree to which the sanctioning structure has to be elaborated or extended to categorize the target of the relevant categorization judgment as, respectively, its elaboration or extension. The parameter of entrenchment, in turn, is (primarily) correlated with the frequency of previous activation of a structure on the one hand, and the likelihood of its subsequent activation on the other hand; with repeated use, a novel structure becomes more and more entrenched so that, eventually, it receives the status of a conventional unit, and thereby is included in the inventory of the symbolic resources of a given language. For the sake of illustration, a sample of recently formed derivatives with -free is given in (2) below; they are put in an order from adjective s which have already become well-entrenched in the grammar of English (sugar-free, virus-free, acid-free) to those which are encountered only occasionally (noise-free, smoke-free, border-free), and yet, with repeated use, they may also gain the status of conventional units of that language.
(2) sugar-free chewing gum; virus-free diskette/program/software; acid-free paper; noise-free mobile-phone; smoke-free city; border-free Europe
Note now that a network of any complex category has a particular internal structure; a lexical network, for example, may center around a given expression's prototypical sense, which functions as a cognitive reference point for categorizing other senses of that expression as instantiations or extensions of the "global" prototype in question. (6) The configuration of the network for privative adjectives, on the other hand, is typical of word formation categories in languages like English, where the derivation of morphologically complex words by means of co-functional affixes is common. In this case, there are two productive co-functional suffixes: -less and -free, and therefore, the overall network subsumes two subnetworks with their respective prototypical patterns. (7)
Before going into differences between prototypical -less and -free patterns, let me first observe that, in terms of cognitive grammar, privative adjectives profile an atemporal relation (a state) which characterizes the entity designated by the modified noun as "being without something" or "not having something" (cf. e.g., a legless soldier a traffic-free zone). (8) Typically, moreover, in its cognitive domain matrix, (9) the profile of a privative adjective co-activates a background conception of the same entity having the thing in question. Observe, however, that the background conception may refer to either a real or a potential/canonical state of the thing designated by the modified noun. For example, the background conception which is evoked by the expression a legless soldier describes a state when a given soldier had his leg/s in reality; by contrast, legless, when said of a child who was born as such, refers to a potential state of that child, or a canonical state in general. A traffic-free zone, in t urn, evokes a background conception of the same zone with traffic in it, while the background conception of acid-free paper refers to what, until very recently, used to be the typical way of making paper in general.
Moreover, since privative adjectives evoke an idea of how the course of events leading from the background state to the profiled state has developed, the specification of their profile requires reference to the cognitive domain called the Dynamic Evolutionary Model (see Langacker 1991: 275-278), which represents our idealized conception of how the world is structured (i.e. the structured world model) on the one hand, and reality evolves (i.e. the elaborated epistemic model) on the other hand) (10) For our immediate concern, it would suffice to briefly characterize two interrelated aspects of this model, namely the conceptions of the "evolutionary momentum of reality" and the "normal course of events". The "evolutionary momentum" refers to the tendency of reality to "continue its evolution along certain paths in preference to others" (Langacker 1991:276), i.e. due to structurally inherent dependencies and connections between events in the world, as soon as the required "enabling conditions' are met, certain ki nds of events are bound to occur under the "normal course of events", i.e. one on which no external force was directed. (11)
Even though both subcategories of privative adjectives refer to the idea of the evolutionary momentum and the normal course of events, they crucially differ in how they portray the way in which reality has evolved towards the state in profile. Specifically, the semantic structures of adjectives with the suffix -less encapsulate an image of the profiled state obtaining as a result of the normal course of events. Moreover, since prototypical adjectives in this subcategory give a highly negative evaluation of the profiled state, they portray the course of events as lying beyond human control. For example, legless, as in a legless soldier, assigns a negative axiological value to the profiled state, implying that if the conceptualizer from whose perspective the image is construed had control over the course of events, (12) he would have rather preserved the soldier intact, i.e. in a positively evaluated state specified in the background conception of the profile. (13) Note, additionally, that these aspects of the meaning of legless are also preserved in this adjective's metaphorically extended sense of 'drunk'.
By contrast, giving a negative evaluation of a state which is evoked in the background, prototypical adjectives in -free (e.g., acid-free, traffic-free, smokefree) take, as their profile, a positive and highly desirable state which would not obtain under the normal course of events. In this case, the background conception against which the profile is characterized can be described in the form of the following complex scenario: since, out of its own driving forces, reality was expected not to evolve towards a state which (from the perspective of a particular person or a group of people) is highly positive and desirable, some human agent/s (some member/s of' that group) must have taken intentional actions to counteract the normal course of events for the desired state in question to have obtained.
The above differences in the background conceptions become especially clear when we look at pairs of expressions like:
(3) a. sugarless chewing gum
b. sugar-free chewing gum
Note, first, that even though chewing gum without sugar can be both sugarless and sugar-free, each of these two expressions is likely to be used in a quite different context. So, we might say that chewing gum is sugarless if we expect or want it to have sugar, and it has not got any. By contrast, chewing gum qualifies as sugar-free if it is intentionally made without sugar because some people do not want to have sugar in it. Moreover, in the first context, the chewing gum would not qualify as sugar-free, while in the other it would not be referred to as sugarless. It is thus evident that prototypical adjectives in -less and -free crucially differ in terms of background assumptions and expectations against which their respective profiles are characterized. And, to put it in more general terms, pairs of privative adjectives as in (3a) and (3b) are specified with respect to the same or very similar cognitive domains ("cognitive base"), and they differ in "imagery", (14) i.e. the way in which a particular concept ual content (or a given expression's cognitive base) is conventionally structured or construed.
Taking the perspective of the non-reductive approach to meaning advocated by cognitive grammar, we should now observe that a speaker's conventional knowledge of the complex category privative adjectives cannot be reduced to a particular constructional schema or schemas. Rather, a cognitive representation of this category should be treated as a full schematic network, wherein each node represents a particular semantic pole of a (specific or schematic) symbolic unit, and together with the phonological pole, characterizes a distinct "semantic variant" of the symbolic unit in question.
Importantly, adopting the usage-based approach, Langacker expects that, depending on their experience and the categorizing judgments they have made, speakers may arrive at different configurations of networks. And, specifically, we have no direct ways to establish how far "upward" individual speakers expand a given network through the process of abstraction (schematization), and in particular, whether they each extract a superschema having all other nodes as its direct or indirect instantiations. Neither do we know how far "downward" speakers spell out their networks into finer grained conceptions. "Speakers may very well differ in these respects, and also in the specific set of categorizing relationships they establish between the nodes" (Langacker 1988: 135). Differences of this kind, however, would pose hardly any problem for communication - provided that the speech-act participants' networks have enough nodes in common.
Observe, further, that, in terms of the usage-based approach to meaning, we should also expect that, as a set of cognitive routines (i.e. units), each particular network is, in fact, a continually evolving dynamic structure. Under the pressures of language use, its configuration is constantly modified -- some nodes lose salience, while other nodes, with repeated use, become progressively more entrenched, and thereby they support the extraction of new subschemas within it. Note also that, on a closer analysis, the categorization by extension and by schema are, in fact, two sides of the same "coin": they constitute two interrelated aspects of a single categorization judgment. To be more precise, any extension which is based on similarity opens up the possibility of establishing a salient schema: the apprehension of similarity between the two structures coincides with the perception of what is common to them, and thereby with the extraction of a schema that both structures instantiate. Should a schematic concept ion of this type become well-entrenched, it would represent a new semantic variant of a given expression or of a grammatical pattern, and as a consequence, a reconfiguration of the relevant network would follow. It is thus evident that "[i]n the final analysis ..., [a schematic network] is not something a speaker has, but rather what he does" (Langacker 1987: 382).
The dynamic aspect of networks is especially clear in the case of the privative adjectives' network which, due to the increasing entrenchment of adjectives in -free and the -free schemas, (15) must have been substantially reconfigured only very recently. As I will argue in the following section, yet another reconfiguration of this network seems to be under way.
3. Towards a cognitive grammar account of recent derivatives in -less
My aim in the present section is to arrive at a cognitive grammar account of recently coined privative adjectives such as those exemplified in (4) below:
(4) a. paperless (office)
b cordless (telephone)
Given a very high productivity of the -free subcategory, we are confronted with the query of why, for example, the expression a paperless office, rather than a paper-free office, has become the preferred option? In term of cognitive grammar, we might give a simple answer and say that this choice reflects the conventional character of construal. Having said this, however, a cognitive grammarian would also ask the question of whether there is any motivation behind the linguistic convention here? Unsurprisingly, a number of possible accounts of this conventional choice might be postulated. On the one hand, it might be argued that adjectives like paperless give a neutral evaluation of the profiled relation (i.e. the profiled state of the nominal trajector), and thereby they qualify as instantiations of the schema for adjectives like moonless and starless, which itself qualifies as an extension of the prototypical -less patterns in that, unlike the latter (cf. section 2 above), it profiles a neutrally evaluated st ate. However, typical contexts in which adjectives like paperless are used suggest that they give a (very) positive evaluation of the profiled state of an entity they modify. (16) For example, a commonly evoked image of a paperless office (or paperless communication) is that of something highly desirable, since it is environment-friendly and cost-efficient. Also, from the point of view of its designers and user, a cordless telephone (or a cordless remote control unit) is far more desirable and better than its old-fashioned variety with a cord attached to it. In brief, since they evaluate the designated state (highly) positively, adjectives like paperless and cordless cannot be treated as instantiations of the moonless schema.
Let us turn, then, to an alternative account. It should be noted at this point that, within the -less subcategory, there is a relatively small group of adjectives (exemplified by the data in (5a-b)) which give a positive evaluation of the profiled relationship. (17)
(5) a. painless, fearless, spotless, selfless (OED: 'not self-centred, unselfish')
b. stainless steel
c. endless problems/task vs. endless (timeless, deathless) prose
Note, first, that adjectives like fearless and painless (cf. (5a)), and the minor subschema whose extraction they are likely to have supported, qualify as extensions of a schema which is instantiated by adjectives like powerless and shameless (cf. also Gorska 1994). This amounts to saying that, for example, fearless can be categorized as an extension of the powerless schema, since, unlike the latter, it profiles a departure from an undesired emotional state (i.e. the state of 'having fear' is negative) to a desired one (the state of 'having no fear' is positive). However, similarly to powerless, the adjective fearless takes it as part of our background assumptions and expectations that emotional states like 'fear' (just like 'power') cannot be intentionally controlled (note, additionally, that *fear-free, just like *power-free, would imply that we can be that way at our will). This means in effect that, with respect to forces that make someone fearless, the conceptualizer is evoked as an off-stage reference p oint who merely reports that the profiled state obtains due to the operation of the normal course of events (i.e. the one on which no external force was directed). Naturally, the same comments would apply to spotless and selfless in (5a).
Seen in this light, the semantic development of the adjective stainless represents a very interesting case (cf. (5b)). According to the OED, the original sense of this adjective, which goes back to the 16th century was: 'without stain, spot or blemish', as in e.g., stainless mirror of the lake; a soft stainless white muslin. Undoubtedly, it was this original sense that motivated the usage of this adjective in the expression stainless steel. For our present concern, it is important to observe that this usage was well-established in the convention of English long before the -free subcategory entered the picture. Recall at this point that with the advent of the -free patterns speakers of English received an option of designating positively evaluated states which result from some intentional actions taken by a human agent. In terms of this recent development of the overall category, there is evident motivation for coining stain-free steel, rather than stainless steel: under the normal course of events, steel is n ot stainless, but rusty and stained. We may assume, therefore, that, when the option of deriving adjectives with -free has been conventionalized, this aspect of the meaning of stainless, which pertains to the notion of intentional action taken to counteract the normal course of events may have gained some salience, and thereby become part of the conventional value of this expression. Had this happened, however, the adjective stainless in its presently considered sense would no longer qualify as an instantiation of the minor painless pattern, but as its extension instead: (18) unlike adjectives which follow the latter pattern, stainless designates a state which lies within human control, and, moreover, obtains due to intentional actions which were meant to counteract the normal course of events. In brief, an original instantiation of the above -less pattern may now have been recategorized as a semantic extension of this very pattern. (19)
The data in (5c), in turn, illustrate a case where the adjective in -less receives a negative or a positive evaluation, depending on the domain evoked by a specific nominal trajector (cf. problems vs. prose). Note, however, that, whichever of these two axiological judgments is imposed by the context, the profiled states are portrayed as obtaining due to the operation of the normal course of events, which lies beyond the human control. When we say, for example, that problems and tasks are endless we are, in fact, admitting that it is beyond our control and power to counteract the normal course of events. Likewise, in terms of our 1CM of how the world is structured, prose normally becomes endless on account of its inherent merits, rather than at somebody's will and desire.
To recapitulate, within the -less category, there is a minor pattern which gives a positive evaluation of the designated relation, yet this pattern agrees with the semantic specification of all the other subschemas for adjectives with the suffix -less in that it evokes an image of the (normal) course of events on which no extemal force was directed. With respect to this generalization, an alternative account of the meaning of adjectives like paperless and cordless (cf. (4a-b)) can now be formulated. Namely, it might be claimed that such adjectives are fully motivated by the above described minor pattern in that they profile a positively evaluated state which obtains under the normal course of events.
It seems, however, this account would tell only a part of the story. Note that there is a clear difference between the motivation for the usage of adjectives like painless (cf. (5a)), on the one hand, and the presently considered new derivatives like paperless and cordless on the other hand. In objective terms, we can hardly control the fact whether we are in the state of, e.g., 'having' or of 'not having' pain, hence the usage painless (rather than painfree) is experientially motivated. By contrast, the state of an office which qualifies as paperless and of a phone which is cordless lie within human control, and moreover, the relevant states are each especially designed by humans, since they have recently come to be evaluated highly positively (of course, from the perspective of particular social groups). Clearly, given this conceptual content, we should, in fact, expect the derivatives paper-free (office) and cord-free (telephone), rather than their newly coined counterparts in -less. (20)
Let us then consider yet another alternative analysis. Suppose we assume that, even though the designated states are not the outcome of the normal course events but of human design, i.e. despite the fact there have been objective grounds for the derivation of the adjectives paperfree and cordfree, the (generalized) conceptualizer has chosen to portray the scene as if the designated state had evolved under the normal course of events. Or, in other words, having derived paperless (office) and cordless (telephone) the conceptualizer has constructed an image of the reality's evolutionary momentum being mature enough to bring about the positive states in question smoothly and naturally. It is crucial to observe at this point that, had the conceptualizer chosen to derive derivatives paperfree and cordfree, he would -- by means of these semantic structures -- have encapsulated a very different assessment of the reality's evolutionary momentum. In accordance with the semantic specification of the -free patterns, (21) the evolutionary momentum would be portrayed as highly unlikely to carry reality in the direction of the profiled state. Consequently, in order to obtain the desired state, some external force would have to be directed to counteract the normal course of events.
And thus, by describing, for example, an office as paperless the conceptualizer gains the option of portraying the designated conceptual content from a vantage point which is different from the viewpoint he would be bound to adopt had he chosen the image encapsulated in the adjective with the suffix -free. By choosing paperless he takes a vantage point of members of a social group who believe that the evolutionary momentum of reality has reached a stage when the normal expected state of an office is to be paperless. By contrast, the corresponding derivative with -free would impose a vantage point of a social group which assumes that, for an office to be paperfree, actions need to be undertaken to counteract the normal course of events.
What, then, is the essence of this recent change? And, in particular: how do the new derivatives in -less differ, on the one hand, from the subgroup of well-established derivatives in -less which also give a positive evaluation of the profiled state (cf. e.g., painless and spotless in (5a)), and the -free patterns on the other hand? Recall that in the case of the subgroup of adjectives like painless and spotless, just like in the case of any well-established -less pattern, the conceptualizer is evoked as an off-stage reference point who is merely reporting that the profiled state obtains due to the normal course of events, or, in other words, due the forces inherent in reality itself. This means in effect that the profiled scene receives "objective" construal. (22) In the case of new -less adjectives such as paperless, on the other hand, the profiled scene becomes more subjective in that the conceptualizer, assuming a particular vantage point, makes an assessment of the likelihood of the evolutionary momentum of reality taking a particular course (to be more precise: the likelihood of the profiled state obtaining under the normal course of events); and, depending on this epistemic judgment, he either chooses the option of the derivative in -free (if the likelihood is assessed as low) or the one in -less (if the likelihood is believed to be high). To put it in terms of the revised view of subjectification, the scene is more subjective since, to say that the conceptualizer makes an assessment of the likelihood of the evolutionary momentum of reality taking a particular course, amounts to saying that the force dynamics is portrayed as inherent in the conceptualizer's mental experience (the conceptualizer's mental extrapolation of reality), rather than in reality itself (see Langacker 1998: 85).
Note, finally, that the recently coined adjectives in -less and -free adjectives (and, by the same token, the -free schemas) can also be said to differ in how they conceive of the locus of force which leads to the positive state in profile. In the case of adjectives with the suffix -free, the locus of force is far more prominent and well-defined, since they evoke the idea of intentional action of a human agent aiming to counteract the normal course of events. By contrast, in the case of newly coined adjectives such as paperless, it is some "generalized force" - the reality's evolutionary momentum which is expected to bring about the positive state in question. By the same token, adjectives in -less of this new type bear a higher degree of subjectification than the -free patterns in that, portraying the force-dynamic relationship as an aspect of the world structure, they are maximally diffuse in regard to the source of potency. (23)
There is no doubt that, under pressures of language use the category of English privative adjectives has been going through diverse changes for some time now. Prior to the increase in entrenchment of new adjectives in -free (with the concomitant effect of establishing -free constructional schemas), English privative adjectives were used to designate states (atemporal relations) of a nominal trajector which have obtained due to the forces inherent in reality itself, i.e. due to the operation of the normal course of events; moreover, conceived against the background of positively evaluated state of a given nominal trajector, the profiled states were most commonly evaluated as negative (cf. e.g., legless, childless). Designation of neutral states was far less common (cf. moonless) and positively evaluated states constituted a minor regularity within this subcategory (cf. e.g., painless, and fn. 17). The conventionalization of constructional schemas in -free marked two important changes in the category. Firstly, positively evaluated and desirable states of the nominal trajector may now be designated by means of privative adjectives which are derived by means of productive patterns in -free, rather than the aforementioned minor schema in -less. And secondly, unlike the latter schema, the patterns in -free can be used to derive adjectives profiling desirable and positive states which are conceived as unlikely to occur due to forces inherent in reality itself, and therefore their occurrence requires that some intentional human action be taken to counteract the normal course of events. As a consequence, the idea of assessment of the likelihood of (the evolutionary momentum of) reality taking a particular course has been encapsulated in the semantic structure of privative adjectives in -free. In brief, in the case of the earlier configuration of the category, i.e. with -less adjectives alone, the conceptualizer was merely reporting about the state which has obtained due to reality's internal forces; having the option of a privative adjective in -free, on the other hand, he can now convey his epistemic judgment of reality taking a particular course.
This development, however, has created a "gap" in the overall category of privative adjectives, since with the help of-free adjectives the likelihood of reality taking a particular course towards a given positive state can be assessed from one kind of perspective only. Namely, a -free derivative may only convey the perspective of those who believe that reality is highly unlikely to reach the profiled state out of its own driving forces. I hope to have shown that, to the extent they can be said to express the opposite assessment, the adjectives of the paperless type may well mark a change which would fill in this "gap".
(1.) I would like to gratefully acknowledge the helpful comments form the participants of a seminar on English privative adjectives which I conducted at the University of Hamburg in November 2000.
(2.) For the theory of cognitive grammar, see, in particular, Langacker (1987, 1991, 2000) and the literature cited therein; see also Gorska (2000).
(3.) A detailed account of the overall network for the category of privative adjectives is given in Gorska (1994), see also Gorska (1995).
(4.) For our immediate purpose, I will restrict the discussion of categorizing relationships to those holding between the semantic poles of symbolic structures. It should be observed, however, that the same type of relations may hold between the phonological poles of such structures, and also between individual symbolic structures of any degree of specificity and complexity (i.e. any kind of a specific expression or a grammatical construction). (4)
(5.) The perception of similarity gives rise to metaphorical extensions whereas the perception of association -- to metonymic extensions. For recent studies into metaphor and metonymy see, in particular, Lakoff and Johnson (1999), and Panther and Radden (1999).
(6.) For more on the internal structure of linguistic categories see, in particular, Langacker (1987: Chp. 10), Taylor (1995).
(7.) It should be observed that the subcategory with -less includes over 600 items, while the one with -free about 40. The difference quite naturally follows from the historical development of privative adjectives: the -less subcategory has been well established in English since the 15th century (see Marchand 1969), while the derivatives in -free are all recent formations and only a portion of them is listed in the dictionaries of English. The -free pattern, however, is constantly gaining productivity: new derivatives are plentiful, and their token frequency increases.
(8.) This generalization does not always apply to scalar variants of privative adjectives, since they may be used to characterize a given entity not only as being without the thing designated by the nominal stem (as in: completely sugarless tea, rainless season, profitless business), but also as not having it to some specified degree (as in: almost/nearly sugarless tea, rainless season, profitless business). Note also that 'thing' is used here in the sense of Langacker's (1987) notional characterization of nouns. To be more precise, it refers to the profile of nominal stems with which the suffixes -less and -free are integrated; the nominal stem functions as the landmark and the modified noun as the trajector of an atemporal relation which is profiled by a given adjective (for more on the profile of adjectives and other atemporal relations see Langacker (1987)).
(9.) A cognitive domain itself can be any sort of conceptualization: a perceptual experience, an abstract notion or an entire knowledge network, and a domain matrix is a set of cognitive domains evoked by a given expression's profile. Most (if not all) cognitive domains are assumed to have the status of an Idealized Cognitive Model (an 1CM).
(10.) Langacker employs these ICMs in the semantic description of both modal verbs and the so called "general validity predications" in English (see Langacker 1991, 1997, 1998, 1999, 2000). It should be noted that these models rest on the idea of reality characterized as "the history of what has happened or been the case, up through the present, as assessed by some conceptualizer" (Langacker 1997:203).
(11.) The need to resort to "our expectations of how the world is structured" in the description of adjectives in -less is also recognized by Holmqvist and Pluciennik (1996). In their study, the authors concentrate on a very intriguing question of why adjectives in-less and-ful sometimes come in pairs and sometimes do not. This question, however, should be taken with caution, since, as Marchand (1969: 262) notes, in the 15th century, when -ful had lost its original meaning 'full', the semantic correlation between the adjective in -less and -ful (harmless-harmful) "practically no longer existed. This does not mean that their semantic opposition had ceased to be felt altogether (in many words it is felt to the present day), but the distance between the suffix was no longer the one that existed between the words full and less".
(12.) By the term "conceptualizer" I mean here some "generalized" or "generic" individual (i.e. in its present sense, this term does not refer to the participants of a speech event). For the notion of a generalized/generic individual in semantic description of lexical items see, for example, Langacker's (1998: 77) analysis of the extended sense of across, as in: Last night there was an altercation right across the street. (12)
(13.) In my earlier account of privative adjectives (Gorska 1994, 1995), I assumed that the semantic pole of such adjectives is also partially motivated via three conceptual metaphors: (BODY) PARTS ARE POSSESSION (e.g., legless, acid-free), RELATIVES (AND PEOPLE CLOSE TO US) ARE POSSESSIONS (e.g., childless, wifeless) ATTRIBUTES (PROPERTIES) ARE POSSESSIONS (e.g., rainless, profitless, smoke-free, traffic-free). However, the presently considered change in the category of privative adjective does not pertain to the metaphorical structuring of such adjectives, and therefore, I will not go into this aspect of their semantic structures any further.
(14.) It should be emphasized that "imagery" in the present sense of the term should not, by any means, be equated with sensory or visual imagery (cf. Langacker (1987: 110ff.); instead, it refers to our basic cognitive ability to "construe a conceived situation in alternate ways -- by means of alternate images -- for purposes of thought or expression" (Langacker 1987: 110). So called "aspects/dimensions of construal/imagery" include, among others, "profiling" (i.e. focusing on the most salient substructure) and "background assumptions and expectations" (for more on this topic see, e.g., Langacker 1987, 2000; Casad 1995). (14)
(15.) See the examples in (2), and also footnotes 3 and 7 above.
(16.) I base the present analysis on the data I have compiled from Time magazine and the British National Corpus (available online: http://sara.natcorp.ox.ac.uk/lookup.html).
(17.) Out of over 600 adjectives in -less, the presently considered subgroup comprises about 20 adjectives.
(18.) For the notions of instantiation and extension see section 2 above.
(19.) Naturally, analogous comments would apply to any other derivative with -less which was well-entrenched in the convention of English prior to the emergence of the -free patterns, provided, of course, the derivative with -free constitutes a reasonable option to the given derivative with -less.
(20.) Note that the case of these adjectives is different from that of stainless (cf. (5b)) in that the latter had been well-entrenched in English prior to the conventionalization of the -free patterns.
(21.) Cf. the semantic import of, e.g., acid-free (paper), focus-free (camera), smoke-free (city), or traffic-free (zone); see also section 2 above.
(22.) According to Langacker, differences in the degree of subjective/objective construal constitute one of the crucial aspects of semantic structures, and an increase in the degree of "subjectification" is a common development of diverse grammatical elements (see Langacker 1991, 1998, 1999). Any detailed discussion of these issues, however, would go beyond the scope of the present article. For the present purpose, suffice it to observe that, in terms of Langacker's revised version o f subjectification, an entity is subjectively construed to the extent that (some aspect of) it "inheres in the process of conceptualization itself, rather than being an onstage object of conception" (see Langacker 2000: 299; cf. also Langacker 1998, 1999). Note further that, in terms of this characterization, in the case of well-established patterns in -less (cf. section 2), the profiled scene is subjective (or subjectified) to the extent that the conceptualizer is evoked as an off-stage reference point who has assigned a particu lar axiological judgment.
(23.) It is interesting to note that, in terms of the proposed account, the changes within the category of privative adjectives are in many ways analogous to the development of modal verbs in English (for the latter see, in particular, Langacker 1991, 1998, 1999, 2000). This topic, however, would deserve a separate study of its own.
Athanasiadou, Angeliki -- Rene Dirven (eds.) 1997 On conditionals again. Amsterdam: John Benjamins.
Blank, Andreas - Peter Koch (eds.)
1999 Historical semantics and cognition. Berlin - New York: Mouton de Gruyter.
Casad, Eugene H.
1995 "Seeing it more than one way", in: John R. Taylor - Robert MacLaury (eds.), 23-49.
1994 "Moonless nights and smoke-free cities, or what can be without what? A cognitive study of privative adjectives in English", Folia Linguistica 28: 413-435.
1995 "Possession frame in word derivation: A case of English privative adjectives", in: Gunnar Persson - Henryk Kardela (eds.), 35-51.
2000 "What's 'cognitive' about cognitive grammar?", Kwartalnik Neofilologiczny 47: 269-279.
Holmqvist, Kenneth - Jaroslaw Pluciennik
1996 "Conceptualized deviations from expected normality. A semantic comparison between lexical items ending in -ful and -less", Nordic Journal of Linguistics 19: 3-33.
Koenig, Jean-Pierre (ed.)
1998 Discourse and cognition: Bridging the gap. Stanford: CSLI Publications.
Lakoff, George - Mark Johnson
1999 Philosophy in the flesh: The embodied mind and its challenge to western thought. New York: Basic Books.
Langacker, Ronald W.
1987 Foundations of cognitive grammar. Vol. 1: Theoretical prerequisites. Stanford: Stanford University Press.
1988 "A usage-based model", in: Brygida Rudzka-Ostyn (ed.), 127-161.
1991 Foundations of cognitive grammar. Vol. 2: Descriptive application. Stanford: Stanford University Press.
1993 "Reference-point constructions", Cognitive Linguistics 4: 1-38.
1997 "Generics and habitual", in: Angeliki Athanasiadou - Rene Dirven (eds.), 191 -222.
1998 "On subjectification and grammaticization", in: Jean-Pierre Koenig (ed.), 71-89.
1999 "Losing control: Grammaticization, subjectification, and transparency", in: Andreas Blank - Peter Koch (eds.), 147-175.
2000 Grammar and conceptualization. Berlin - New York: Mouton de Gruyter.
1969 The categories and types of Present-Day English word formation. Munchen: C.H. Beck'sche Verlagsbuchhandlung.
Panther, Klaus-Uwe - Gunter Radden (eds.)
1999 Metonymy in language and thought. Amsterdam: John Benjamins.
Persson, Gunnar - Henryk Kardela (eds.)
1995 New trends in semantics and lexicography. Umea: Acta Universitatis Umensis.
Rudzka-Ostyn, Brygida (ed.)
1988 Topics in cognitive linguistics. Amsterdam: John Benjamins.
Taylor, John R.
1995 Linguistic categorization. Prototypes in linguistic theory. (2nd edition.) Oxford: Clarendon Press.
Taylor, John - Robert MacLaury (eds.)
1995 Language and the cognitive construal of the world. Berlin - New York: Mouton de Gruyter.
|Printer friendly Cite/link Email Feedback|
|Publication:||Studia Anglica Posnaniensia: international review of English Studies|
|Date:||Jan 1, 2001|
|Previous Article:||Composite Predicates and idiomatisation in Middle English: a Corpus-based approach (1).|
|Next Article:||A corpus-based study of the figure and ground in siting, standing, and lying construction.|