Ontology as a View on Truth

David Bozarth

Sonoma State University

Philosophy 302

4 July 2004

 

In Ethics without Ontology, Hilary Putnam inveighs against warmed-over Platonism – an ontological perspective on truth – creeping into our ideas of what is known. He says that ontology has "outlived its usefulness" by posing "a single 'literal' sense of 'identity' - one which is cast in marble, and cannot be either contracted or expanded."1

Wittgenstein derided the claim that "all general assertions, including ethical and mathematical ones, can be said to 'describe reality.’”2 Putnam labels this type of claim an "inflationary ontological view" (p17) that presupposes the existence of “non-natural [abstract] objects” in the tradition of the Ideal Forms of Plato (e.g., Truth, Justice, and the Good Life). In ethical reflection, recognition of the limitations of Platonic idealism began with Aristotle (p19), who sought the value of a given principle in the particular qualities of its manifestation in the apparent world.3 We will see how inflationary, as well as reductive and eliminative ontological conceits fail, under Putnam’s scrutiny, to form a basis for ethical discourse.

 

Objectivity Without Objects

According to Putnam, there is a world of truths that are not descriptions of real objects.

ANALYTIC truths are a priori truths; those that are truly self-evident. “Up is up” within any frame of reference where “up” is defined, irrespective of any other fact or concern.

SIMPLE TAUTOLOGY is truism, not a product of reflection.4 An example of simple tautology, or simple logical truth:  p => (p Ú q). We don’t replace the variables p and q with anything. The tautology is purely formal, and may be used routinely in evaluating more complex logical statements.

CONCEPTUAL truth is not analytic, and does not describe a real object. It is “truth by general agreement,” relative to current understanding, and is therefore “objective” truth. In principle, conceptual truth is subject to revision. The defining criterion for a conceptually true statement: We can’t make sense by asserting its negation. (Because conceptual truth is revisable, though, it’s conceivable that someday the assertion of its negation could make sense, in light of new understanding.) (p60-64)

There are some categories of conceptual truth; and some ways to confuse them with, or distinguish them from, descriptions.

            Complex logical truths. These are non-obvious tautologies, not necessarily very complex, but they must be either accepted as axiomatic, or proved by accepted methods, before their truth may be agreed upon – thus their truth is relative to current understanding, and in principle subject to revision. Like mathematical and simple logical statements, they are purely formal and do not describe real objects. A logical statement of this type is

((p => q) ^ (q => r) ^ q) => (p Ú r).5

            Mathematical truths. Non-trivial mathematical statements must be proved, like complex logical statements. These proofs draw on logic and on fields of accepted mathematical understanding, subject to revision. Pure mathematical statements do not describe objects that really exist, as Putnam establishes with two observations.

First, there is generally no clear identity relation for mathematical objects – consider that a set may be defined as a function, and any function may be defined as a set. A strong mathematical statement has general applicability to various classes of object.

Second, mathematical “existence” almost always means “possibility of existence,” and nothing else. Putnam objects to taking “exists” as only a “manner of speaking” which translates to “hand-waving”. He also objects to taking “exists” in the sense as used in quantificational logic (“’For all X’ implies ‘there exists an X’”), where the quantifying expression indicates representation of a set of real objects. If one insists on a quantificational meaning for “exists”, and admits that such set-objects don’t really exist, then one is back to hand-waving; otherwise one must accept ontological robustness as a quality of such non-natural (abstract) objects – a Platonic “existence” for mathematical “objects”. 6  (p79-81)

Natural language expressions of symbolic logical forms. "Up is up" is analytic, as is "Down is down". "Up is up and down is down" is a conceptual truth, an expression of our current (and possibly permanent) understanding that the pure conjunction of two true statements yields a true statement.

Such a statement is not, in general, a description of real objects: a valid natural language statement form, is not valid because of any quality of the objects referenced by the statement components. Its validity comes from the logical characteristics among the statement components. In fact, the inference may be valid whether or not the component statements are true. For example, any natural-language rendition of  (p => q) ó (~q => ~p) is true. (p59)

            Methodological value judgments. Is a scientific, cultural, or creative theory simple, coherent, plausible? Why? Do those who succeed with such assessments possess a rare gift of conversation with an abstract realm – or are they just skilled and dedicated? For Paul A.M. Dirac, who received the Nobel Prize in Physics in 1933 at the age of 31, the “beautiful” theory was always to be preferred over the “ugly”. (p68)

Putnam suggests that rather than being a form of special intuition about "non-natural objects", this kind of evaluation is a tool-in-trade of science, art, and design. Though some aptitude could be innate, the skill of field-enlightened evaluation is best developed through training and practice. The process embraces complexity and seeks elegance - like finding a best path through an unmapped forest, based on prior knowledge of forests, trees, and trails. Putnam calls this process "non-deductive" reason, closer to complex logic than to prevailing notions of intuition.

 

Objectivity of Ethics

Some ethical truths may be descriptions, but not all. The statement "Terrorism is criminal" is not a description; rather it is a simple moral evaluation, according to Putnam. (p73) On the other hand, the statement "Terrorism in Iraq involves beheadings" describes an actual situation (a real object). The implied moral evaluation is secondary to the action of describing.

Putnam maintains that “certain crucial ethical statements” are conceptual truths, not descriptions, and "denial of ... objective judgment in any of the cases in which we have objectivity without objects has disastrous consequences." (p77-78) Let us examine some consequences of the varieties of ontological error, as posed by Putnam.

            Inflation. If all objectively true statements are descriptions of real objects, then any ethical statement that is not a description cannot be true, and so has no value. (p52-53) Thus when we encounter an ethical statement of principle with abundant and enduring value (say Kant’s Categorical Imperative), we face three distinct choices:

1) Find an “object” (“that maxim”) and claim that the Imperative “describes” the maxim.

2) Determine that the statement is untrue, and the principle is without value.

3) Wave the hand.

Of these choices, the first merits consideration. A maxim has no mass and no location, so it must be a non-natural object of the abstract realm. Putnam maintains that to posit a non-natural object of Platonic character is unsatisfactory.  He calls this a “pseudo-explanation” (p78) – one that fulfills a unique need and does no useful work beyond justifying the point at hand. (p60)

Statements, moreover, are themselves not real objects of description; therefore compound ethical statements that logically relate simpler statements are not descriptions of objects. "The idea … that a statement of logical consequence is a description of an intangible object or set of objects, is utterly empty as an explanation." (p60)

            Reduction. Putnam typifies this view as “’A is nothing but B’ … for example, … ‘ethical utterances are nothing but expressions of feeling’…” (p19-20) The reductive ontologist claims that a statement is true because of its correspondence with a logically valid structure. An ethical truth, then, is merely a dressed-up tautology.

But evaluating ethical reasoning by correspondence with symbolic logic is a failed enterprise, according to Putnam. (p78) Valid logical structure is necessary, but is not a sufficient condition for a natural language statement to represent ethical truth. One can sidestep by claiming that the tautology must be well-dressed with true and honorable premises – but this is not sufficient for mapping a true natural-language statement to a tautology,7 nor is it necessary for mapping validity from a tautology to a natural-language statement,8 nor is it sufficient for doing so.9

A problem is that logical validity and conceptual truth are not identical; the validity of a logical expression is not the same “kind of truth” as the truth of an instance of its natural-language expression. In this spirit, Putnam takes issue with the philosopher Quine’s definition of tautology: “an instance of a schema all of whose substitution instances are true".  An example is the simple tautology  p => (p Ú q). But we have no example of a natural language schema (e.g., "subject + verb + object") that is true in all instances. (p57)

Ethical truth, then, is not only a function of the structure of its language. The objectivity of ethical discourse is not contingent on supposed identity of its truth with a valid logical structure (i.e., its objectivity is not “ontologically committed” to logical validity). Ethical statements and arguments containing them embody objectivity when, in addition to being structurally sound, they meet commonly acceptable, reasonable standards of relevance and continuity.10

            Elimination. Where the reductionist says, “You’re confusing A with B,” the ontological eliminationist says, “A doesn’t really exist.” (p20) An inflationary view of truth as description, as we have seen, forces a choice between Platonism and elimination. In refusing the Platonic ideal while accepting inflationary ontology, one must deny that ethical reasoning can be objective, on the ground that no objectively true ethical statements are possible.

In this manner the study of ethics can be relegated to psychology, sociology, and cultural anthropology. Primal urges and subconscious drives can be regulated, after all, by the systematic discipline of society. Powerful tools of data collection and analysis can assist theorists in determining which behaviors and trends are prominent, meaningful, promising, or pathological. The efforts of ethicists can be directed away from “reflection” on presumed “truths” and toward the development of functional-behavioral cyberstructure to help stabilize the empire.

The consequences of denying objectivity to conceptual reasoning extend beyond philosophy, though. (p78) Ethical statements are a special case of all those that lose validity in the process of inflating the standard of truth while sincerely de-Platonizing objectivity. As shown above, methodological value assessment relies on conceptual judgment. We could no longer trust the Nobel laureate who claims value in beauty, because beauty would be entirely subjective. How about experimentalists – what criteria would they use to fashion their data sets?

Similarly, conceptual reasoning – that is, refined judgment based on field knowledge – as applied in law, medicine, education, the arts, technology, government, cooking, auto-racing, dating, and channel-changing would be left to flounder in subjectivity.

Invalidating conceptual judgment does have potential benefits, though. Difficult passages in Kant, for example, in belonging (as a consequence of Quine's definitions) to a "second-grade [conceptual] system" less formalizable than science, fail to have "ontological significance" - that is, such passages merely fail to exist. (p83-84) Many students could find this reassuring.

 

Ethical discourse

Ethical reasoning is practical reasoning; thus ethical judgments are "as fully governed by norms of truth and validity as any other form of cognitive activity," because these standards "are internal to practical reasoning itself." (p72)

If ethical valuation is a rational process with definable methods that are themselves subject to standards and evaluation, then why are many ethical claims controversial?

According to Putnam, even questions of fact are not always so prone to agreement as is commonly supposed. A "general feature of practical problem resolution" is the involvement of "factual estimates on which it is difficult if not impossible to ever get convergence." In Putnam’s example, the question of whether it’s possible for a democratic, peaceful, secure, fully socialist society to survive and prosper is likely to remain an open question for the foreseeable future. (p76-77) In principle, the question is empirically resolvable, but prevailing powerful interests, led by multinational business, most likely would not allow such an experiment to proceed in any case.

Conversely, ethical questions are not all so inherently controversial. Ethical questions sometimes involve context of relevant belief. The indeterminate nature of some such ethical questions may improperly be translated to a presumed "irreconcilable" quality for all ethical questions. Debates on the morality of abortion, for example, tend to center around questions of when the fetus becomes a person, or acquires a soul. The “person” question belongs to biology, psychology, and law; the “soul” to theology and metaphysics. Unless such key questions were to be resolved, the morality of abortion would remain debatable. This is not a general statement about all ethical questions, though. (p75)

Controversy over ethical valuations also stems from the conceptual nature of the relevant reasoning. Perfect “general agreement” is difficult to find on the questions underpinning some ethical concerns, and this is not always due to belief or other “subjective” factors. As Putnam shows, conceptual judgments are objective statements about ideas or relationships, and conceptual truths are conceptual judgments that can be established on the basis of generally accepted understanding - but not every conceptual statement is true.