New Foundations of Ontology.
This is Bergmann's last book, published posthumously. It is not mainly methodological or meta-ontological, as the term "foundations" in the title might suggest. Rather, it expounds Bergmann's late ontology which contains several important and wide-ranging innovations. One could get glimpses of these from papers published around 1980. In the book Bergmann develops his late ontology in full and in great detail.
Bergmann's middle ontology, which can be gathered from his books Logic and Reality (1965) and Realism (1967), has the categories of things, facts and subsistents. The things fall into the subcategories of universals and bare particulars. The most important among the subsistents are the ties which tie things into facts and simpler facts into more complex facts. Things are considered as simple and facts as complexes of things (if atomic) or of facts (if molecular). Facts are the only complexes, since subsistents are simple, too. Including universals as a subcategory, Bergmann's category of things is obviously different from the ordinary concept of things. Ordinary things are categorised by Bergmann as facts. His particulars are taken to be constituents of ordinary things which individuate them.
The major categorial innovation of Bergmann's late ontology is the appearance of another kind of complex (the circumstance) and of a category taken to stand between simples and complexes (the class). The circumstance differs from a fact in not being tied together by a tie. The constituents of a circumstance form a complex eo ipso, as Bergmann says. What is categorised as a circumstance in the late ontology was categorised as a fact in the middle ontology, namely diversity and sameness, the intentional connection between mental act and object, and the elementhood connection between each element of a class and the class. Classes which constitute a category of their own in the late ontology had been categorised as things, as derived universals, in the middle ontology.
The introduction of tieless complexes is an unexpected change in Bergmann's ontological development because he earlier took the tie to be the key to complexity. Bergmann's conception of complexity has two components: 1) the picture of binding things into a complex, and 2) the contrast between sentences and terms, where sentences represent complexes, terms simples. Classes which cannot be so expressed are considered to be non-complex.
Not only does Bergmann introduce circumstances as a new kind of complex in addition to facts, he also transforms the analysis of facts in such a way that the core of every fact is a circumstance and the unity even of facts is tieless. The entities which were taken to be ties in the middle ontology become functions which "cling" to the core. The reason Bergmann gives for dropping the concept of tie is that it is anthropomorphic. His concept of function, which replaces that of tie, is distinguished from Frege's and traced back to Meinong's "objects of higher order".
The circumstances which form the cores of facts are diversities. Thus, in the simple case of a particular exemplifying a monadic universal the core of the fact is the diversity of We particular and the universal to which the exemplification function then clings. Diversity holds always between two entities. It is a dyad, as Bergmann says. Hence, a fact with more than two constituents has to be based on iterated diversity, on diversities of diversities etc., on dyads of dyads etc.
This iteration of diversity is not particularly plausible and the related treatment of facts makes them very different from the way they present themselves to perception or other mental acts. Bergmann is aware of this difficulty. He calls it "phenomenological distance", i.e. distance from the phenomenological data. But the new treatment of facts allows him to solve a problem which he did not notice before and which could not be solved by his middle ontology, the problem of order. The task is to give an ontological account of the order of relata in a relational fact and of the order of constituent facts in a molecular fact. Bergmann's solution, according to which e.g. the order of two relata is founded on the diversity of one of them from the diversity of the two, is inspired by Kuratowski's set theoretic explication of the ordered pair.
Molecular facts (conjunctions, disjunctions, negations, general and existential facts) can be very complicated. There are conjunctions of disjunctions of negations of general facts etc. But although they occur already in his middle ontology, Bergmann did not pay much attention to their structure and form. Now, the iterated dyads of diversity allow a systemic reconstruction or standardisation, as he calls it, of all facts and circumstances. Bergmann lays down explicit rules of this stepwise reconstruction in which the functions (of exemplification, of conjunction, etc.) loom large. These rules are expressed by conditional statements of existence saying that a certain complex exists if a certain dyad and a certain function exist. Bergmann stresses, however, that these rules (or in his term "canons") do not exist as such and that their ontological ground is nothing but the existence of the respective function. This, as well as the wording of the canons, is presumably designed to make clear that though we have to reconstruct complexes stepwise there is no temporal and deliberate process of construction corresponding to it.
Not only are Bergmann's complexes not products of a synthesising agent (as in much of the modem tradition, e.g. in Kant), he also suspects that thinking of complexes as containing ties is a leftover from the anthropomorphic production conception of complexity. However, Bergmann's final treatment of complexes may seem to be no more satisfactory. It suggests that it is diversity which holds the constituents of a complex together, since the dyad in a fact contains all its constituents (the function not being taken to be a constituent) and is a complex independently of the function. That diversity is offered as a ground of unity seems strange. We normally conceive of diversity as separating. But that is merely a metaphorical conception of diversity, in which it is thought of by analogy with spatial separation. Naturally, diversity is a relationship between the diverse entities and thus connects them. A greater difficulty than that of grounding unity on diversity is again the phenomenological distance. In Bergmann's ontological standardisation of the atomic fact that a is F, a and F form a dyad to which the function of exemplifications clings. But, as the fact presents itself, exemplification is not added to the unity of a and F, but is rather the content or way of this unity. Thus, the view of the middle ontology that exemplification ties a and F into a fact is nearer to the phenomenological data.
Already in Bergmann's middle ontology all complexes have a mode which is not taken to be a constituent but to permeate it, as he now says. There are two alternative modes: actuality and potentiality. A fact with the mode of actuality is what is usually considered to be an actual and existent fact, while a fact with the mode of potentiality to be a merely possible and non-existent fact. However, in Bergmann's middle and late ontology facts with the mode of potentiality exist as much as those with the modes of actuality. They differ only with respect to the mode which permeates them and not with respect to existence. Bergmann gives two main reasons for acknowledging existent potential facts. First, to offer an ontological ground for the well-formedness of false but well-formed and meaningful sentences. The ground offered is that they correspond to potential facts, while sentences which are not well-formed do not correspond to any facts. Bergmann's second reason is connected with his treatment of intentionality (of the relation between a mental act and its object, its intention) which is based on the principle that every act has an intention, even a false one. Bergmann's Principle of Presentation, according to which what is presented, the intention of a mental act of whatever kind exists, reminds one of Parmenides. It derives not only from a certain conception of intentionality but also of existence. Existence is assimilated in the late ontology to the possibility of being presented and no longer acknowledged as an entity as in the middle ontology.
When Bergmann originally introduced the modes of complexes in his middle ontology he held that we are never presented with the mode of a complex. That he thus postulated entities was a difficulty because it conflicted with his phenomenological view binding ontology to what is presented. In his late ontology Bergmann escapes the difficulty by claiming that we are sometimes presented with modes, namely if the complex is analytic. Bergmann thus not only removes a difficulty but also reaches a new explication of analyticity. A complex is analytic if and only if it presents us with its mode. Bergmann's concept of analyticity differs from the customary one by applying to complexes rather than to sentences, and it has a wider extension including, in addition to the complexes corresponding to logically true sentences of the propositional and the functional calculus, all circumstances, i.e. all diversities, elementhoods, and intentional connections.
As already mentioned, there is a changed conception of classes in Bergmann's late ontology. Bergmann gave up the categorisation of classes as derived universals because he dropped this category altogether. He realised that it led to an implausible multiplication of facts. In the late ontology classes form a category of their own. They are built on selectors which are circumstances specifying the conditions for the elements of the class. The class of a and b, for example, is built on the condition of being either the same as a or the same as b. If the class is infinite the selector has to be of a different type. It has to be recursive. Bergmann deals extensively with the problem of classes. But he clearly distinguishes between the set-theoretic and the ontological approach. The usual set-theoretic conception of classes as collections (in Bergmann's terminology) is ontologically unsatisfactory. However, that does not get set theory into difficulties as long as it does not make ontological claims.
The final chapter of the book is entitled: "The Linguistic Turn Contained". The linguistic turn (a common phrase originally coined by Bergmann) in philosophy was taken by Bergmann, too. He started as and continued to be an ideal language philosopher, emulating Russell and the early Wittgenstein. However, for him the formalised language is merely a means of ontological investigation and explication. Too often the linguistic turn led to the substitution of philosophy by (more or less self-made) linguistics and by mathematical logic. That is why Bergmann wants to see the linguistic turn contained. He concludes his book by showing carefully and in great detail that ontology is an independent discipline different from the study of language and the study of calculi.
In Bergmann's late ontology there are entities which are neither simple nor complex, the classes. The simples (the things) turn out to be Two-in-Ones uniting two entities even if these are totally inseparable and have the lowest ontological status. The combination of a complex and its mode is also conceived as a Two-in-One. The basic principle that complexes are different if they have different constituents no longer holds. Quantified facts have several different analyses into constituents. Some facts have no constituents at all (their corresponding sentences consist of logical signs only). Bergmann is well aware of this strain but he rightly sees no alternative to analysis in ontology. And he is not the man to give up so soon as his fellow country man and fellow logical atomist also from Vienna. Compared with its beginnings, Bergmann's ontology has become rather complicated. This is the result of his epistemological realism and his taking ontology seriously as a theory of the world. It is also the result of an increasing grasp of ontological problems and solutions and their connections. New Foundations does not deal with the complete set of traditional problems, as the earlier Realism did, but concentrates on the more modem ones. It is a piece of superb craftsmanship, a rare thing in ontology. These two books make me agree with Hector-Neri Castaneda, who is cited on the blurb as calling Bergmann "the major ontologist of recent decades". New Foundations is difficult and requires careful study. Hence the long introduction and glossary by the editor, William Heald, are very appropriate.