The birth of objects in the C-I component
A restrictive interpretive apparatus suggested above (simplified also for expository reasons, as its being monomodal throughout, which does not affect the point to be made, though) exhibits modesty expected on the part of an apparatus which remains guided by the syntactic component and because of that is unwilling to incur more ontological commitments than necessary—it stands to reason that an influence of narrow syntax makes the part of the C-I component responsible for the immediate processing of syntactic objects ‘have a taste for desert landscapes,’ to use the famous phrase of Quine (1948: 4), although not because it has a taste for specific answers to ontological worries. It does not abhor points of evaluation, domains assigned to them, connecting their members with counterpart relations, using relative accessibility and so forth. But the web of relationships established by counterpart relations contrasts with semantic apparatuses which employ individuating functions (as in Hintikka’s approach) or even introduce functions into the object language subject to quantification instead of elements of the domain (a familiar move in formal semantics, see recently Abels and Marti (2010, 2012), van Urk (2015) for defenses of this solution). The latter approach makes such semantic objects part of the ontology tout court; ontological costs following adopting the former one are lessened by the fact that individuating functions are not considered part of the object language: Hintikka (1969a) goes so far as to claim that they are therefore not part of ontology altogether (a similar claim is made e. g. in Hintikka (1969b: 39)):
It is true that by Quine’s criterion they would seem to be part of our ontology, since we have to quantify over them. This impression is misleading, however, and in fact brings out a clear-cut and important failure of Quine’s dictum, construed as a criterion of ontological commitment. The functions in question are not inhabitants of any possible world;
they are not part of the furniture of our actual world or of any (other) possible world. Thus it would be extremely misleading to count them in in any census of one’s ontology. What they represent is, rather, an objectively given supply of ways in which we can deal with more than one contingency (possible world). They are part of our conceptual repertoire or our ideology (in something like Quine’s sense) rather than part of our ontology. In a sense, we are committed to their existence, in the sense of their objectivity, but not to including them among ‘what there is’ in the actual world or in any other world. (Hintikka 1969a: 179)
Even if such claims are to be accepted—although they are hardly uncontroversial, given that such functions are quantified over indeed, ordinary members of domains not serving the purpose appropriately (see also Tselishchev (1977, 1979) for some discussion of the point)—it is a remarkable property of models as envisaged in the preceding discussion that they do not incorporate such intensional objects and are therefore not ‘in any sense committed to their existence.’ True enough, the C-I component may further construct models which would provide individuating functions fulfilling roles prescribed by Hintikka; and it is still far away from having ‘prefabricated objects’ as members of domains of points of evaluation. Rejection of the latter idea may be rightly considered to be one of major improvements over the standard way of approaching modal logic that Hintikka’s proposal introduced, together with the idea that objects as we know them are rather to be considered constructed in interpretation than given in advance; already accepting this move would give formal semantics possibilities to maneuver its apparatus so as to avoid several charges raised against the referentialist stance as a foundation for semantics of natural language. It is worth seeing Hintikka argue for the superiority of this view on domains:
The best way out for Carnap would have been the one he did not take, namely, to give up the one-domain assumption, and with it the priority of individuals over possible worlds. Ironically, it looks as if such an idea of individuals as constituted rather than given ones was in much more agreement with Carnap’s own line of thought in the Aufbau than with the one-domain assumption. In any case, if an individual is simply whatever is picked out by an individuating function, there is no a priori question as to how such individuals can be combined with each other. All such questions will be a posteriori ones, and Quine’s criticisms become redundant. Of course, there is a price to pay, for there is no such thing as a free lunch among logicians, either. What happens is that we cannot make any a priori assumptions about the behavior of the individuating functions or of the “world lines” of cross-identification which they serve to define. These lines may fail to be extendible from one world to another in the most radical sense imaginable, and we cannot, sight unseen, rule out branching and merging among them. I remember pointing out this idea once to Richard Montague. His characteristic response was: “But that would complicate our logic enormously”. The right rejoinder of course would have been:
“So much the worse for your logic”. (Hintikka 1992: 177-178)
The price to be paid—that ‘we cannot make any a priori assumptions about the behavior of the individuating functions or of the “world lines” of cross-identification which they serve to define’—if considered from the point of view of accounting for interpretive procedures applied by the C-I component to syntactic objects is rather a gain than a loss, exonerating the interpretive apparatus from a charge of importing a priori assumptions directly into the interpretive process; the price of ‘complicating logic enormously’ should not be an obstacle, Montague’s reaction (predictable indeed, see section 3.1) notwithstanding. It might be noted that viewing objects as constructed, as theoretical posits, which for scientific reasons may be dissolved without loss, is Quine’s view:
Reference to things in space-time (...) could be reconstrued as reference to the sets of number quadruples comprising the space-time coordinates of the things. So far as evidence goes, objects figure only as neutral nodes in the logical structure of our total theory of the world. (Quine 1993: 112)
Needless to mention, Quine’s views on the matter (one may compare also Quine (1992b: 9); see further Keskinen (2010, 2012, 2013) for discussion of Quine’s views on objects) are part of the whole web of interrelated claims about learning, reference, and evolution of cognitive faculties lying far away from a minimalist stance on such isses, developed and modified over years (see Quine (1960, 1973, 1992a, 1995) for particularly extensive discussions of relevant points); nor does his avowedly extensionalist position allow him to be content with Hintikka’s intended answers to Quinean doubts about legitimacy into opaque contexts and rejection of the device of cross-world identifications (see Quine (1976, 1998)). What is common is that epistemological concerns have priority over ontological ones: ontological issues come to be investigated, questions posed and answers established with epistemology providing a guidance. If this hierarchy of priorities and the very ‘idea of individuals as constituted rather than given ones’ reminds of Kantian and Husserlian lines of thought, it should, for it is in this direction that Hintikka’s proposals lead; in the words of Smith (1983), ‘the motivation is Kantian: individuals are “synthesized” in consciousness according to rules or principles ingredient in our human conceptual system (...) In Kantian and Husserlian terms, what is at issue is the “constitution” of individuals’ (Smith 1983: 261).
Embracing a rapprochement between phenomenology and philosophically oriented research in intensional logics—the latter appropriating and reconceptualizing several notions and ideas of the former (see F0llesdal (1969, 1990) for seminal discussions and Smith and McIntyre (1982) for a full development of the framework), not without voices of disagreement and dissent, but also attempts at reconciliation (see e. g. Mohanty (1981), Hintikka (1981), Harvey (1986), Hutcheson (1987), Harvey and Hintikka (1991), Overgaard (2008))—Hintikka claims that world lines as they are understood on his analysis are ontologically innocent (‘Ontology is a matter of‘the furniture of the world (...) Such questions are not affected by the constitution of world lines,’ (Hintikka 1975: 218)), and, rejecting Lewis’s counterpart-theoretic approach as it came with the whole package of metaphysical assumptions incompatible with his own stance (see section 3.2.1) and tying the cognitive activity on the part of a subject with the emergence of world lines, stresses the fact that ‘they are neither here nor there’:
Speaking of ‘constitution’ easily evokes idealistic associations. (...) No doubts are thrown by possible-worlds semantics on the reality of the actual world or of its inhabitants (...) Constitution is not a domestic matter, as it were, but a matter of foreign policy, that is to say, a matter of cross-world comparisons. What is brought about in constitution is literally neither here nor there (...) Constitution does not create inhabitants of any possible world, only methods of comparing entities in different possible worlds for their identity. (Hintikka 1975: 216)
Proceeding as advertised above, the C-I component is upon meeting syntactic objects yet less involved in epistomology-related questions than it would be if it adopted individuating functions, which, to be sure, may later arise due to its cognitive involvement with other modules: relationships that it establishes between elements of domains with the help of counterpart relations and the apparatus of points of evaluation are not only as neutral ontologically and epistemologically as possible, they are also extremely local in nature, confined to bounds set by discontinuous syntactic objects—chains. Whether such local counterpart lines may be extended further and provide basis for individuating functions is not the matter for the C-I component at this stage to decide nor to worry about. It is not concerned with such interpretive properties as modal constancy or modal separation (in the sense of Belnap and Muller (2014b)), either, although it may be assumed to avail itself—beside a procedure employing counterpart relations tout court—also of an enriched procedure with a requirement that relevant counterparts satisfy ‘local constancy’ along the modal path that is traversed during evaluation of a chain (an addition which seems plausible in the realm of A-bar movement), such properties being possibly established at further stages of its cognitive activity (note that we also skip over complications arising with counterpart relations connecting tuples of objects, see Hazen (1977, 1979), Lewis (1983: 44-45), Kupffer (2000/2010) for different attempts at a solution of problems with such cases). What narrow syntax and the C-I component as following in the footsteps of the former provide are bricks and stones for ‘the human “co- gnoscitive powers" which provide us with rich means to refer to the outside world from intricate perspectives’ (Chomsky 2010: 57). Doing so, they become responsible for our human mindedness.