Sunday, October 26, 2008

Towards a Lutheran Theological Semantics III

Imagine two theories T1 and T2 indiscernible with respect to their syntax. To give an interpretation to this syntax is to define an ordered pair <I, n> such that I specifies a domain D of entities named by the individual constants of the theory, some Fx specifies a subset of D, and Fx . . . k specifies a k-ary Cartesian product in D. Let n now designate a naming function from names in the language to D, monadic predicates to subsets of D, and k-ary predicates to k-tuples in D. The function thus assigns for each and every nonlogical symbol an extension in D. What we are doing here is assigning a semantics to our language. Obviously, if both T1 and Ts use the same >, they will mean the same thing. Two theories indiscernible with respect to syntax and having the same interpretation have the same model. We say that M models a theory T if and only if all the sentences of the theory are true given the projection of the language onto the model. Obviously, if M models T1, and T1 and T2 have the same interpretation and mapping function n, then M shall model T2 as well.

Within the practice of science, the syntax of theories change as a function of new empircal data and concomittant theory adjustment. In science generally, the method of projection of the syntax of the theory upon a model is for the most part invariant, and it is this invariancy that makes possible scientific progress generally. Words like 'electron', 'boson', and 'p orbital' retain their interpretation (reference) across different theories generally. (We might say that in a situation of revolutionary change in paradigm, new interpretations and naming functions might arise.)

However, within the practice of theology, things are far different. Scripture and theological tradition has worked to produce a rather loose 'theological theory' whose syntax does not in general change. But as times change, the syntax of this theological theory takes on a new interpretation. Imagine T1 being classical christological formutions at the Council of Chalcedon in 451, and T2 be the same classical christological formulations said by Paul Tillich in 1957. Here it is obvious that while the syntax of T1 and T2 remains the same, there is a change of the mode of projection of the syntax upon its model. Because T1 and T2 are both regarded as true, there are distinct models M1 and M2 that model the same syntax. The same syntax is modeled both by M1 and M2, or alternatively, M1 and M2 both model the same theory T1. (Remember that T1 = T2 syntactically.) The situation now is that we have two distinct models for the same syntax, two distinct ways that the world might be ordered that would make possible the truth of T1 = T2.

The question is this: What is the theological theory T1 and is it different then from T2 after all? The answer is, of course, that we do not regard scientific theory as mere syntax, but as syntax + an interpretation. Similarly, I aver, we ought not to regard theological theory as mere syntax but syntax + an interpretation. How can it be then that many today in theology, particularly in ecumenical theology, think that syntax alone does a theology make? How can it be that the Joint Declaration on the Doctrine of Justification can claim agreement on 'justification' because the syntax of the language is similar between Lutherans and Rome?

We should remember that Luther said he was not interested in agreement in words (verbis) but in things (in rebus). Although Luther was not using the language, he was, of course, interested in the disparate models the same theological syntax could sustain. When one thinks about it, this is how it has always been in theology. Was this not precisely what happened after Ephesus (431) that two sides used the same words while allowing different interpretations of the same language?

A theology that has lost interest in its interpretation and naming function, is a theology that has lost interest in truth, because only with the assignment of models is truth put in play. While sytax deals with form and structure, semantics deals with truth and meaning. Theology has always been about the latter. It is a mark of the recent theological poverty of our time that we could have been so bewitched for so long, and have not even noticed.




Wednesday, October 08, 2008

Towards a Lutheran Theological Semantics II

Being realist with regard to a class of theological statements does not mean one has to commit to metaphysical realism, to a claim that there are self-identifying theological objects, properties, relations, events or states of affairs that obtain apart from human awareness, conceptualization and language. One might allow some ontological contribution on the part of the subject, and still claim that the principle of bivalence holds for theological assertions, that these assertions are either true or false (but not both), and that their truth-conditions are evidence transcendent, i.e., that although, in principle, adequate evidence cannot be found for the truth of statements like 'God was in Christ reconciling the world to Himself', such statements are true or false solely on the basis of whether the conditions do in fact obtain.

But the question immediately arises, can Lutheran theology be realist in this sense? Is not Lutheran theology committed to a bundle of paradoxes that become problematic if theological statements are interpreted in a realist way? Can a Lutheran who wants to assert that humans are justified and sinful simultaneously, that the bread really is bread but is the body at the same time, that God is both love and wrath simultaneiously honestly embrace semantic realism. If theological statements are either true or false, then A has a definite truth value (true or false), and if ~A is simultaneously asserted, then A & ~A hold simultaneously, and from a contradiction anything follows whatsover. So is not semantic realism incompatible with Lutheran theological assertion? Must not the assertions of theology be given a different analysis, perhaps as projetions of human emotion, disposition, or existential orientation? If Lutheran theology is commited to paradoxical claims, must it not reject semantic realism, for to accept it is itself incompatible with the assertion of those paradoxical claims?

This question goes to the heart of the problem. If there are paradoxes that must be asserted, and such paradoxes are vicious in and for any ontic divine domain, then to assert those paradoxes must entail that the assertions are not really assertions about an ontic divine domain; they are expressions of, or assertions about, that region of being for which the law of excluded middle is relaxed: human being. Human being is a paradoxical reality, it seems, because it is a being, who in its being has its be-ing at issue, a be-ing where both the projective possibilities of the future, and the remembered facticities of the past are simultaneously consistutive of its present. If I am both who I am (because of who I was), and who I am (because of who I might be) then I am both who I am and who I might be. Since I might be different than I was, then I am both A & ~A simultaneously. From the fact that my present is constituted by both my past and future, I am both who I am (I was) and who I am not (who I might be who is not who I was). The fact that I am both simultaneously gives the Lutheran hope, it seems, to think the simul eschatologically. I really am sinful because I have always been so. At the same time, I am justified because the possibility of being no longer sinful is my possibilty given the life, death and resurrection of the Christ. I am thus both who I am (sinful) and who I am not (sinless on account of Christ) at the same time. Similar moves must be made for the other Lutheran paradoxes; language about objective paradoxes must be traced back to the existentiality of the subject, to the being of that being who lives simultaneously as thrown pro-ject. It is no small wonder why Lutheran theology should find solace in Kierkegaard, Nietzsche and Heidegger. It seems, to paraphrase Tillich, that existentialism really is the good luck of Lutheran theology.

But much is lost with this approach. To justify paradox in theology by tracing the paradoxes back to the subject makes theological assertions essentially autobiographical. Talking about God cannot, as Barth said, be achieved by talking about ourselves in a loud voice. Giving a semantic realist construal to theological statements is to assert again what Luther and the Reformers presupposed: that God existed independently of us, and that statements about God are true or false if the truth-conditions of those statements are met. 'God is in Christ reconciling the world to Himself' if and only if God is in Christ reconciling the world to Himself'. The meaning of the statement is given by its truth-conditions, by specifying those things that must be so if the statement is true. So how can Lutheran theology embrace seeming paradox and still accept semantic realism? The answer to this question, I believe, drives Lutheran theology back into its Catholic origins. Lutheran theology did not spring forth from Catholicism as a separate theological school with its fundamental terms and statements logically disconnected from, or (worse yet) incommensurable with, the fundamental terms and statements of late medieval scholasticism. Lutheran theology is Catholic theology with a few twists. It emerged and grew in the waters of a late medieval scholasticism that was nominalist in its ontology and realist in its semantics. But that the Lutherans could argue that a justified person can remain sinful suggested that something quite ontologically different was at work in Lutheran theological thinking than had been thought before. Coming to an honest appraisal of this difference can help us, I think, in our quest for answering the question: How is a Lutheran realist semantics possible?

Tuesday, October 07, 2008

Towards a Lutheran Theological Semantics I

While syntax deals with the form and structure of a language, semantics deals with its meaning and truth. When considering the general question of the semantics of a language, one must get clear on the meaning of the terms of that language, and the meaning and truth of sentences comprised of those terms. Terms having meaning apart from other linguistic units were called by the medievals categorematic terms. Terms not possessing meaning on their own (e.g, 'is', 'of') were called syncategorematic terms. In establishing the semantics of a language, one has to specify the meaning of a class of primative terms, and then show how the meaning of those terms contribute to the meaning of sentences in which those terms are ingredient.

In ascribing a semantics to theological sentences, one routinely examines their truth-conditions. The truth-conditions of a statement are those conditions which must obtain if the statement is to be true. For instance, the truth conditions of 'the cat is on the map' are those conditions which must obtain if 'the cat is on the mat' is true. These conditions simply are the state of affairs of there being a cat, a mat, and the relation of the cat being on the mat. Simply put, 'the cat is on the mat' is true if and only if the cat is on the mat.

Now in considering these truth-conditions, the question arises as whether or not one can specify evidence transcending truth-conditions. While it is in principle possible to have a perceptual causal connection to cats sitting on mats such that one can truly know when or if the requisite conditions are fulfilled such that 'the cat is on the mat' is true, this is not so with regard to most theological statements. What causal connection can one have to states of affairs like 'there are three distinct persons united in one divine being or essence'? How could one ever be said to know this is the case? Those who for this reason reject the possibility of evidence transcending truth-conditions are antirealist with respect particular classes of statements. While language like 'there are three distinct persons in one divine being or essence' are perhaps warrantably assertible, they have no truth-conditions, for the condition for the possibility of truth-conditions cannot be met. To accept the possibilty of evidence transcending truth-conditions for a class of statements is to be realist with regard to that class of statements. The question is this: Should Lutherans accept a realist construal for their theological language? Over the last 200 years Lutherans have progressively become less and less sanguine that theological language has truth-conditions, and that is thus can be given a realist construal.