Halliday & Matthiessen (1999: 565):
… we are offering the ideation base as a conceptual alternative [to] the mind, knowledge, cognition … the concerns of cognitive science.
… we are offering the ideation base as a conceptual alternative [to] the mind, knowledge, cognition … the concerns of cognitive science.
The ideation base, by dint of being polysystemic, accommodates variation along this cline, not only from folk to scientific but also across alternatives: it embodies both congruent and metaphorical construals of experience, and it provides elasticity within the overall construction space.
As a resource for making sense of experience, the ideation base enables us to construe a range of different theories, commonsense as well as scientific. There is a cline between folk, or commonsense, theories and scientific, or uncommonsense, ones; and at any point along the cline alternative theories may be in competition.
To say that the ideational system is polysystemic means that it can support these different theoretical angles on experience: semantic variation of all kinds is the manifestation of the different theoretical interpretations that language places on experience.
From an educational point of view, the most fundamental complementarity is the move from the registers of everyday life to the registers of education: this is a move from folk or commonsense models to “uncommonsense” models of systematic and technical knowledge.
Now, just as the overall ideation base is a theory of our total experience of the world around us and inside us — the theory that is shared by the culture as a whole, so also the different registerial variants constitute different ‘subtheories’ of our experience. These ‘subtheories’ may complement one another by simply being concerned with different domains of experience … This complementarity is purely additive, although for society as a whole it constitutes the semiotic aspect of the division of labour, whereby different people construe different facets of the overall cultural experience. But such subtheories may also be concerned with more or less the same domain, bringing alternative perspectives on the construal of experience that is shared.
In addition, the ideation base is polysystemic in another sense: registerial variation. We have seen that such variation can be construed in terms of the probabilistic nature of the linguistic system, as variation in the probabilities associated with terms in systems. Seen in this light, a register is a particular probabilistic setting of the system; and the move from one register to another is a re-setting of these probabilities. What is globally the ‘same’ ideational semantic system can thus appear as a collection of different systems, as one [i.e. our viewpoint not the speaker] moves along the cline of instantiation from potential to instance… . As we noted above, the effect is quantitative; but it is also qualitative, in the sense that it provides different perspectives on experience within the same system.
Such complementarities [metafunctional, fractal, systemic and metaphorical] constitute one form of indeterminacy of the system — one that allows it to be "polysystemic" in the particular sense of embodying more than one way of construing experience.
… beyond its congruent mode, there is a metaphorical complementarity: the ideational model offers a complementarity between the congruent mode itself and the metaphorical mode, making it possible to take some phenomenon as already construed and then reconstrue it as if it was a phenomenon of a different kind.
… still in the congruent mode, there are systemic complementarities: the ideational potential offers systemic complementarities such as the ergative and transitive models of participation in processes, and the mass and count (singular/plural) models of quantity …
… within this congruent mode, there is a fractal complementarity: the highly generalised semantic types of projection & expansion are manifested in complementary domains — those of sequences, figures, and participants; so that, for example, some phenomenon of experience construed as having temporal expansion might appear either as a sequence or a configuration …
… in the congruent mode of construing experience, there is a metafunctional complementarity: the ideational potential offers two complementary modes for construing experience — the highly generalised logical mode, with projection & expansion as the dominant semantic motifs, and the more particularised experiential mode, with its typology of processes, things, qualities, and circumstances …
The ideational meaning potential embodies not one single semantic system but rather several such systems coexisting; in Firth’s terms, it is a “system of systems” — in two distinct but related ways [internal complementarities and registerial variation].
It seems likely that all these different kinds of indeterminacy are what make it possible for the grammar to offer a plausible construal of experience — one that is rich enough, yet fluid enough, for human beings to live with. We should stress once again that the examples cited here are features of the ideation base of one particular language, namely English. No other language will be identical. Indeed the distribution of indeterminacies is likely to be precisely one of the features in which languages differ most, and even perhaps varieties within one and the same language. But every language depends on indeterminacy as a resource for meaning — even if our grammatics is not yet very clever at teasing it out.
There are of course many different contexts for all these indeterminacies, in different regions of the total semantic space. Certain types of ambiguity appear to be not so much artefacts of the realisation (not just grammatical puns, so to speak) but rather another kind of complementarity, where the grammar is as it were "having things both ways" — both interpretations have to be accepted at one and the same time. This is sometimes the case with Token + Value structures, in figures of being. These clauses are always ambiguous, if the verb is be, since this verb does not mark the passive; yet some depend on being interpreted both ways — particularly, perhaps, some proverbial sayings, Thus, one man's meat is another man's poison is both Token ^ Value 'what one person likes may displease another' and Value ^ Token 'what one person dislikes may please another'; contrast what's sauce for the goose is sauce for the gander, which can be interpreted only as Token ^ Value.
An intermediate degree of specificity, with partial neutralisation, can be seen in the non-finite clause with accompanying preposition, as in they get caught for taking bribes. What happens here is that the fact that there is a connection between the two figures is unequivocally construed by the dependency; but the nature of this connection — what kind of logical relationship is being set up — does not enter the picture.
When two figures are linked into a sequence, by some logical semantic relation, there is a rather wide range of possible semantic relations between the two: the relations of time and of cause and condition are particularly elaborated in this respect, but there are others besides — the manner, the matter, and so on. The distinctions among these relationships, however, may be to a greater or lesser degree neutralised, where one clause is construed as dependent on the other; this happens as the dependent clause moves from finite to non-finite status. For example, in they get caught taking bribes the distinction that would be made in the agnate finite clause, among, say they get caught if they take bribes, they get caught when they take bribes, they get caught because they take bribes, is simply neutralised — it is not a blend of all three, nor is there any ambiguity involved.
In those figures where there is a second direct participant, some form of agency runs through all the different types of process; but agency is such a complex aspect of human experience that the grammar does not delineate it by a single stroke, but construes it by means of a fundamental complementarity, that between the transitive and the ergative perspectives. Thus figures involving two direct participants, such as Actor + Goal in the material, are aligned along two different axes: the transitive one, based on the potential extension of force (mechanical energy) from a doer to another entity; and the ergative one, based on the potential introduction of agency (causal energy) from another entity as external source. Thus the earthquake shook the house is construed both as 'earthquake + shake' plus optional Goal 'house’, and as 'house + shake' plus optional Agent 'earthquake’. As always in cases of complementarity, certain parts of the region are more strongly aligned to one perspective, other parts to the other, but the total picture requires the confrontation of the two.
… the grammar distinguishes a number of types of process, material, mental, verbal and relational; the distinctions are made by a cluster of syntactic variables… . But since these variables "draw the line" at different places, there are areas of overlap, with mixed categories that share some characteristics with one group and some with another. We gave the example of behavioural processes; these are a mixed category, formed by the overlap of the material, on the one side, and the mental or verbal on the other. Behaving is construed as a type of figure that (like the mental) typically has a conscious participant as the central rôle, and does not extend beyond this to a second participant; but, on the other hand, it does not project, and it has a time frame like that of the material. Thus behavioural processes lie squarely athwart a fuzzy borderline.
… the grammar distinguishes a number of types of process, material, mental, verbal and relational; the distinctions are made by a cluster of syntactic variables —
the valency of associated participant rôles,
the class of entity that takes on each rôle,
the potential for combining with other figures,
the associated tense systems and the like.
But at the opposite corner, so to speak, if we combine low value with oblique (remote), the result is typically blending rather than ambiguity: e.g. it couldn't hurt you to apologise is a blend of 'it would not be able to hurt you' (readiness: ability), 'it is unlikely that it would hurt you' (probability) and even perhaps 'it would not be allowed to hurt you' (obligation). In other words, looking at it from the point of view of blending, in the region of 'what I think’/ 'what is wanted', it is easiest to blend the low values 'what I can conceive of’ with 'what is permitted', especially in 'remote' conditions (hypothetical, projected or tentative) [realised as could, might]; and hardest to blend the high values 'what I am convinced of’ with 'what is required', especially when 'immediate' [realised as must].
There is a great deal of indeterminacy throughout the region; but it is of more than one kind. At one "corner", if we combine value with neutral (immediate), then the resulting wordings are ambiguous, as to the type of modality expressed: thus must has three clearly distinct meanings, (a) as probability (e.g. that must be Mary 'certainly that is Mary'), (b) as obligation (e.g. you must wear a helmet 'it is essential that you wear a helmet'), (c) as readiness: inclination (e.g. if you must make all that noise 'if you insist on making all that noise’). That these are truly ambiguous can be gathered from an example such as she must complain, which has to be interpreted in one or another of these different meanings — the context will of course usually make it clear which;
Modality, the speaker's angle on what is or what should be, is notoriously fluid and shifting in its categories, probably in every language. In English, there is a fairly clearly defined semantic region construed at the intersection of a number of grammatical systems, including (1) type: probability / usuality // obligation / readiness: inclination / ability; (2) value: median // high / low; (3) orientation: objective / subjective; (4) immediacy: immediate (neutral) / remote (oblique); (5) polarity: positive / negative; and one or two others. These are realised synthetically in various ways, one of which is by the modal finite operators can, could, may, might, will, would, should, must, ought-to (and one or two other fringe members).
But, thirdly, within the overall construction of experience, the diversity of spheres of social action is realised by variation in the line-up of semantic features — that is, by variation in register. The probabilities are reset; and in some cases one or two "critical systems" are strongly affected in this way, such that the local norm skews the system, or perhaps even reverses the skewing set up by the global norm. It is here that we find future taking over as the unmarked primary tense in weather forecasting. As we said above, we define register variation in just these terms, as the ongoing resetting of probabilities in the lexicogrammar, which then functions to construe the ongoing variation at the level of the social process.
Secondly, systems have varying probability profiles, so that (in terms of information theory) they carry differential loads of information: the skewer the probabilities of the terms in a system, the greater the redundancy that it carries — hence the less we need to attend to its unmarked state.
How would we summarise the significance of indeterminacy, from the point of view of the semantic construal of experience? It lies first and foremost, perhaps, in the general principle that is being proclaimed, if indeterminacy is a typical and unremarkable feature of the grammar: that 'this is the way things are’. Our “reality” is inherently messy; it would be hard to construe experience, in a way that was beneficial to survival, with a semiotic system whose typical categories were well-defined, clearly bounded, and ordered by certainty rather than probability. This is the problem with designed systems, including semiotic ones: as a rule, they fail to provide adequately for mess.
… when choice is made in two systems simultaneously, such that each serves as the environment for the other, there is often a conditioning effect on the probabilities. This may be an indication of a change in progress, or it may be a stable feature of the overall system.
Synchronically (that is, viewed synoptically in this way as a meaning potential) a language is, as we have said, a probabilistic system: if we say that, in the grammar, there is a system of primary tenses past/ present/ future, we assume the rider 'with a certain probability attached to them’. But we do not, of course, speak or write with one grammatical system at a time. Systems intersect with each other simultaneously (we choose tense along with voice, polarity, mood, transitivity and so on), and they follow each other in linear succession (we choose tense in clause 1, again in clause 2, again in clause 3 and so on). Each instance has its environment, both of previous instances, and of simultaneous instances of systems with their own sets of probabilities.
At the “instance” end, a single highly–valued instance may exert a disproportionate effect: quotations from the Bible and from Shakespeare are familiar triggers of this “Hamlet factor” in English … But such qualitative effects take place against a background of microscopic quantitative pressures, the sort of nanosemiotic processes by which a language is ongoingly restructured as potential out of the innumerable instantial encounters of daily life — the “sheer weight of numbers”, as we sometimes call it.