From Metaphysics, Back to Semantics

Given its centrality in the semantic framework in which it was articulated, it is no wonder that the notion of being took on so many varieties and gave rise to so many metaphysical problems. After all, in this framework, all true predications are taken to be verified by the actual being of the significata of our predicates, whether those significata are taken to be the natures or substantial forms of substances, or their accidental forms, their quantities, qualities, relations, actions, passions, times, places, positions or habits, or even not any genuine forms, but rather the lack thereof, that is, privations, or other beings of reason, such as negations or relations of reason. In fact, we may add to the list of such ‘‘quasi-entities’’ demanded by this framework the significata of entire propositions, variously called dicta, enuntiabilia, or complexe significabilia, etc. (Nuchelmans 1973, 1980).

In general, one may say that in this framework there is some ‘‘nonchalance’’ toward admitting semantic values (as they called them, significata and supposita, cf. Read 2008) of our expressions in our ontology, relegating the task of sorting them out to metaphysics. And this is precisely what thirteenth-century metaphysicians do: they make various classifications of various sorts of beings (based primarily on Aristotle’s system of the categories), and ever more refined distinctions and identifications among the semantic values of various expressions in different linguistic categories, leading to a bourgeoning ontology of entities and quasi-entities, having different degrees of reality (or unreality, for that matter), and different, sometimes rather obscure criteria of identity or distinctness (or quasi-identity and quasi-distinctness: see the above-mentioned examples of Scotus’ less-than-numerical-unity, or Henry of Ghent’s intentional distinction, let alone Scotus’ formal distinction, or Suarez’s modal distinction) (see Suarez 1947).

The following table provides a general overview of the varieties of being entertained by medieval philosophers working in this framework, sketching out ‘‘the big picture’’ of the ontology or ‘‘quasi-ontology’’ (containing also various sorts of quasi-entities) of what may be termed the framework of the realist ‘‘via antiqua semantics’’ in contrast to the nominalist ‘‘via moderna semantics,’’ emerging in the works of William Ockham and his followers, such as John Buridan, Albert of Saxony, Marsilius of Inghen, Peter of Ailly or Gabriel Biel (Table 1).

Given the immense complexity of the resulting ontology, one can certainly understand Ockham’s complaint, according to which his contemporaries are guilty of‘‘multiplying beings according to the multiplicity of terms... which, however, is erroneous and leads far away from the truth’’ (Ockham 1974:169; see also page 171, where Ockham explicitly claims that this is the root (radix) of the errors of ‘‘the moderns’’).

To be sure, Ockham’s charge is not exactly justified. After all, as we could see, even earlier authors could reduce the ontological commitment of their theories by either identifying the semantic values of expressions across different categories (as when following Aristotle, they declare that action and passion are the same motion, or when they identify relations with their foundations in the categories of quality or quantity, etc.) or by assigning them some ‘‘diminished’’ ontological status, by relegating them to some class of ‘‘quasi-entities.’’ However, it is true that earlier authors, taking it for granted that our meaningful words are meaningful because they express concepts whereby we conceive of something, never worried about assigning semantic values to our phrases, and took it to be

Note that I am using the designations of "via antiqua semantics''and ''via moderna semantics'' slightly anachronistically, to designate two radically different ways of construing semantic theory in late-medieval philosophy and theology, which I take to be the conceptual basis for the historically somewhat later separation of the two viae as the altogether different ways of doing philosophy and theology, leading to the so-called Wegestreit, the historical ''quarrel of the ways'' in the fifteenth and sixteenth centuries. For a detailed historical discussion of the late-medieval separation of the via antiqua and the via moderna, see Moore 1989. For the impact of the separation of these two ways on the emergence of''the battle of the faiths,'' Glaubenskampf, in the age of the Protestant Reformation, see Oberman 1977.

A metaphysical task to sort out the ontological status, identity, and distinctness of these semantic values, leading to the ever more rarified metaphysical and theological questions in this framework.

Ockham’s real innovation, therefore, consisted not so much in producing a simpler ontology (since that, in principle, would have been available to his predecessors as well), but rather in reining in any unwanted “ontological excesses’’ already in his semantics. That is to say, the real novelty in Ockham’s approach is the elimination of unwanted ontological commitments, not only through metaphysical argument (resulting in identifying the semantic values of different expressions across categories or relegating them to some diminished ontological status), but rather through semantic analysis, which has remained ever since the most powerful weapon in the conceptual arsenal of nominalist philosophers (cf. Goodman and Quine 1947).

Just by way of a quick illustration, let us consider a simple example of how commitment to such quasientities as privations comes about in the older framework, and how Ockham can get rid of it through logical analysis (cf. Klima 1993). Take the privative term ‘‘blind’’, which can obviously be true of something only if the thing in question lacks sight, that it could and ought to have by nature (namely, an animal that ought to have sight, but lacks it for some reason or other). In the older framework, if the predication ‘‘Jerry is blind’’ is true, then this is so, just as in the case of every other true predication, because what the predicate signifies in the subject is actually there in the subject (which is precisely the actuality that the word ‘‘is’’ signifies). But precisely because for blindness to be there is for the corresponding sight not to be there, the being of blindness cannot be taken to be being in the same sense as the being of sight; therefore, blindness has to be a being in a different sense, a being of reason. And similar considerations apply to all sorts of privations, negations, or relations of reason, thus opening up the ‘‘quasi-ontological’’ realm of beings of reason in this framework.

Ockham, on the other hand, uses logical analysis to show that for the true predication of the term ‘‘blind’’ and similar privative terms we need not posit any such quasientities. As the nominal definition of the term shows, this syntactically simple term must have a complex concept corresponding to it in the mind, namely, the concept properly expressed by the nominal definition ‘‘animal not having sight.’’ Now, in this phrase, there are two absolute terms (terms applying equally and indifferently to whatever they signify, on account ofbeing subordinated to the corresponding absolute concepts), namely, ‘‘animal’’ and ‘‘sight.’’ There is also a connotative term (signifying some of the things it signifies primarily and others secondarily, connoting them only in relation to the former, on account of being subordinated to a connotative concept), namely, the term ‘‘having’’, primarily signifying the things that have something and connoting the things had by them. Finally, there is the syncategorematic term ‘‘not,’’ which does not signify anything in external reality, only the mental concept of negation, which is a mental act, merely modifying the representative function of the categorematic concept with which it is construed (cf. Klima 2006). Thus, the complex connotative concept to which the term ‘‘blind’’ is subordinated will signify animals, while connoting their sight; however, on account of the implied negation it will only apply to animals that do not have sight. Thus, on this analysis, the sentence ‘‘Jerry is blind’’ will be true just in case Jerry is one of the animals that do not have sight, just as it should be. However, the important difference of this analysis from the former is that this analysis does not require the actuality of some quasi-entity, namely, the privation of Jerry’s sight: the term ‘‘blind’’ applies to Jerry simply because he does not have sight, which is just what the negative connotation of the term requires. But this negative connotation is effected through the term’s signifying only positive, real entities, namely, animals, their sights, and a positive quality of the mind, namely, the syncategorematic concept of negation.

Without going into further details, perhaps, even this brief illustration will suffice to indicate how, through such and similar analyses, Ockham is eventually able to come up with a much simpler, reduced ontological picture (see Klima 1999). In his semantics, all our linguistic expressions are mapped onto an ontology containing only two really distinct categories of entities, namely, entities belonging to the category of substance, and entities pertaining to the category of quality, some of which, however, are qualities ofthe mind, namely, those naturally representative mental concepts whereby we conceive of objects of our cognition (i. e., the categorematic concepts that provide the sense of our categorematic terms) or those mental concepts that merely have the function of modifying the representative function of the former (i. e., our syncategorematic concepts, providing the meaning of our syncategorematic terms).

As we can see, the semantic construction of the via antiqua certainly demanded the mapping of language onto a rich and complex structure of semantic values, accorded with different degrees or modes of reality, expressed perhaps in correspondingly distinguished senses of ‘‘being’’ (ens), among those who endorsed a strong doctrine of the analogy of being, such as Aquinas and his followers, or without such distinctions, among those who endorsed the doctrine of the univocity of ‘‘being,’’ such as Scotus and his followers. In any case, the rather ‘‘cavalier’’ assignment of semantic values to all linguistic items in any syntactical category in this semantic framework left it to metaphysics to sort out into which ontological categories these semantic values fall, and whether they are distinct or identical. Indeed, the latter questions led to distinctions among various forms of distinctions as well, ranging from numerical and real, to less-than-numerical, intentional, formal, modal, or mere conceptual distinctions. Nevertheless, the metaphysical strategies of distinguishing various modes of being and/or senses of‘‘being’’ and allowing the identification or quasi-identification ofsemantic values of expressions in various categories could in principle yield in this framework a real ontology (i. e., an ontology of real entities, not considering any quasi-entities) no more profligate than the parsimonious nominalist ontology of Ockham, acknowledging only two distinct categories of singular entities, namely, substances and qualities.

Ockham’s genuine innovation, therefore, was not so much in ontology per se, but rather in the ways and strategies he applied to reduce ontological commitment in the semantic theory itself. Thus, for him, the reduction of ontological commitment, in contrast to the via antiqua strategy, was not a separate metaphysical task to bring metaphysical order to a burgeoning quasi-ontology of ‘‘recklessly’’ assigned semantic values, but rather the direct task of semantic theory itself, carefully working out the mappings of distinct linguistic categories onto a parsimoniously construed ontology, containing only real entities in the ‘‘permitted’’ categories.

Thus, Ockham’s ontology can be represented by means of the following, much simpler diagram, cutting the former diagram in half, and reducing its inner contents as well (Table 2).

In this diagram, quasi-entities are gone: the phrases apparently designating them are all analyzed in terms of their nominal definitions, containing only phrases that denote or connote entities in the ‘‘permitted’’ distinct categories of substance and quality. A similar strategy allows the reduction of really distinct real categories from ten to just two. In fact, Ockham’s approach is so successful that one wonders why he even needs two categories. In principle, a simple homogenous category, the category of substance, would do for the purposes of his semantics.

As Marilyn Adams has convincingly argued (notably, on the basis of a passage from Buridan’s Questions on Aristotle’s De anima, see Adams 1989:283-285), the need to posit a distinct category of quality stems from the need to explain genuine alteration as a change in quality, without having to resort to a ‘‘quantitative’’ analysis of it in terms ofan atomistic physics. Thus, although logical analysis alone would allow a reduction of ontology to a single category, further considerations in metaphysics and physics do necessitate further ontological distinctions.

Indeed, it is precisely such further, metaphysical and physical reasons that prompt the great ‘‘systematizer and legitmizer’’ of Ockhamist nominalism, John Buridan to depart from Ockham’s ontology, despite sharing virtually the same semantics (see the entry on John Buridan in this volume).

To be sure, despite the nominalist project of‘‘ontolog-ical reduction,’’ in the nominalist semantic framework itseLf there is still what contemporary philosophers would identify as “quantification over non-existents,’’ and hence ontological commitment to some ‘‘weird entities.’’ However, in general, in medieval logic, quantification was not regarded as carrying ontological commitment. What we can refer to and quantify over may have been something, could be something, or, possibly, will be something, but if it is not actually one among the things presently populating our universe, then it is not anything, i. e., it is literally nothing (cf. Klima 2001:197-226, 2009:143-174). Thus, in the nominalist logical framework to be is simply to be identical with one of the actually existing things. Accordingly, although the affirmative copula in this framework still carries ‘‘existential import,’’ it does not predicate existence (whether absolutely or as qualified by the significata of the predicate). Rather, it is merely a syncategorematic sign of the identity of the supposita of the terms of an affirmative proposition. Therefore, perhaps paradoxically, the verb ‘‘est’’, functioning as a copula, does not even express the notion of being. In this sense, nominalist logic completely broke the strong links assumed in the older framework between the various modes of predication (construed as various ways of saying that something is), and the various ways in which things are.

See also: > Albert of Saxony > Essence and Existence