Page 512 - Encyclopedia of Philosophy of Language

Key Figures
particularly the syncategorematic words—occurring in a syllogism are insufficiently analyzed. For instance, the Fallacie Parvipontane (an anonymous treatise from the late twelfth century) offers the following example of a fallacy of division: "Those two men cease to be; if anyone ceases to be, he dies; therefore those two men die.' The fallacy consists in the fact that in the first premise the word 'cease' is said of the pair of men, but in the second of only one man, so that the conclusion fails in claiming that each of those men dies when only one of them dies (see Kretzmann 1976:105- 106).
In the thirteenth century, William of Sherwood, Nicholas of Paris (fl.1250) and others identified the semantics of syncategorematic words as a specific set of topics and discussed them in separate treatises. In the fourteenth century these discussions were ab- sorbed into the sophismata-literature, which tried to explain paradoxes caused by the confusing use of syn- categorematic words (see Read 1993). The strategies for dissolving the paradoxes may be recognized as an attempt to analyze by linguistic means problems associated not only with logic but also with meta- physics or natural philosophy. For instance, as regards the sophisma 'Socrates is infinitely whiter than Plato begins to be white'—an example discussed by Richard Kilvington (ca.1305-61)—an analysis of the syncategorematic words 'infinitely' and 'begins' was supposed to make understandable the comparison of
a definite degree of a quality (white) in one substance (Socrates) with the state of another substance (Plato) considered at the instant at which its acquisition of that quality begins (see Kretzmann & Kretzmann 1990:153-55).
The analysis of the use of syncategorematic words was linked to an investigation of the functions or properties of categorematic terms. In the logica vetus Anselm of Canterbury distinguished two main func- tions: the appellatio of a term is the actual indication orthenamingofitsreferents,whereasthe significatio conveys the understanding or concept (intellectus) expressed by the definition of the term. Thus, the term 'man' names (appellaf) all individual human beings, but it signifies the concept of a rational, mortal animal. This important distinction was used by Peter Abelard in order to explain that a term can be used significantly even if there is no referent. In winter, Abelard writes, when there are no roses, the term 'rose' has no appel- latio but it nevertheless preserves its significatio, since the understanding of the concept rose does not depend on the present existence of individual roses (see Rijk 1967:190-99).
Gilbert of Poitiers also adopted the distinction between appellatio and significatio, but with respect to nouns—one kind of categorematic term—he claimed that two aspects of signification are to be taken into account: a noun conveys not only an understanding of a thing as such (idquodest) but also of the essential
quality by virtue of which it is this particular thing (id quo esf). Thus, 'man' signifies not only a particular human being, but also the humanity by virtue of which every man is a man, and 'stone' signifies not only a particular stone, but also the 'stone-ness' by virtue of which every stone is a stone (see Rijk 1987:161). This claim is influenced by a Platonic ontology which posits universal beings with real, distinct existence in addition to particular beings.
The logica moderna provided a more detailed sem- antic theory avoiding such strong Platonic claims. This theory recognized four properties of terms (see Maieru 1972: 47-317; Spade 1982): (a) 'signification' is the capacity of a term to bring something to the mind, i.e., to establish an understanding of a thing. This property applies to all categorematic terms; (b) 'supposition' is the reference of a categorematic term or its 'standing for' (stare pro) something, applying only to subjects and predicates (many Continental logicians of the thirteenth century attributed it exclus- ively to substantives) and therefore determined by the actual use of a term in a syntactical context; (c) 'copulation' is the referential function of a dependent categorematic term (an adjective or a participle) linked with a substantive; and (d) 'appellation' is the reference of a categorematic term to an actually exist- ing thing.
Obviously, one cannot attach the same importance to all four properties, since copulation and appellation are subordinate to supposition; copulation applies to a subclass of all suppositing terms, and appellation is a temporal restriction of supposition. For that reason, fourteenth-century authors focused almost exclusively on signification and supposition. These two properties provide the basis for a theory that is not purely logico- semantic, since signification—the establishment of an understanding—is an epistemologico-psychological property of a term.
Logicians in the thirteenth and fourteenth centuries distinguished three main sorts of supposition (for a schematic presentation, see Maieru 1972:306-317). First, a subject or predicate term can stand for some- thing of which it is truly predicable (in most instances an extramental thing), e.g., 'Socrates' in 'Socrates is a man.' The subject standing for an individual human being is said to have 'personal supposition.' Second, a subject or predicate term can also stand for a spoken or written expression, e.g., 'man' in 'Man is a mono- syllable.' In this case, the subject is said to have 'material supposition.' Andfinally,a subject or predi- cate term can stand for a universal, e.g., 'man' in 'Man is a species,' where the subject is said to have 'simple supposition.' This third kind of supposition appar- ently presupposes the existence of universals and therefore has strong ontological commitments. It was redefined by nominalist authors of the fourteenth cen- tury, who claimed that the simple supposition of a term is confined to its reference to a concept (see Michon 1994: 213-44).
490