Medieval religious philosophy, by contrast, disparages anthropomorphic description.⁽⁴⁾ While not quite an abstract entity, God is described, even defined, in abstract terms. The flavor of this is perhaps best conveyed by an example from the Christian philosophic tradition, St. Anselm's characterization of God as "the most perfect being." This is no mere honorific supplement to the anthropomorphic characterizations. It is a definition, one that subsumes specific divine perfections.

My aim here is to explore this tension with an eye to the fate or ultimate status of the doctrinal output of the philosophers. I will argue that theological doctrine⁽⁵⁾ is not a natural tool for thinking about biblical/rabbinic Judaism. The "system of thought" model applies to biblical/rabbinic Judaism only with the application of force.

As a preliminary, let me distinguish two aspects of Talmudic discussion: Halacha, the legal component, and Aggada, the non-legal. The legal discussions of the Talmud are often interrupted with remarks on ethical or spiritual matters, homiletic comments, exegetical discussions, stories about the lives of the Rabbis as well as of ordinary folk, parables, and the like. These aggadic passages are less authoritative than the legal discussions. This is not to make little of them, or to diminish their significance for the religious life.

As a related preliminary, let me distinguish my anti-doctrine approach from another with which it might be confused. It is sometimes said that in Judaism all that matters is practice, Halacha, conformity with the law. It is clear that Judaism is not possessed of an official theology. Indeed, the variety of ways of thinking about God in Jewish tradition is striking. Begin with the highly anthropomorphic picture of the Bible and Aggada.⁽⁶⁾ Move to the medieval rationalist conception, Neo-Platonist and/or Aristotelian. God is, as it were, an unmoved mover; God's transcendence gets top billing, and--to oversimplify--his immanence in the world somehow takes care of itself. Now turn to the Kabbalah, the mystical tradition, with its quite distinctive--sometimes super-immanent--modes of characterization.⁽⁷⁾

It does not follow, however, from the lack of a canonical theology that conformity with traditional practice is enough, all that matters. Practice is central, and not only in that there is a more or less canonical practice.⁽⁸⁾ How one thinks about theological things, or even whether one thinks much about such things, is clearly less important than how one lives in the world. Still, questions of relative importance aside, it would seem to be a mere shadow of Jewish religious life merely to observe the practices, with no attendant mode of religious thought and feeling. I am skeptical about theological doctrine, but this is not to say that practice, Halacha, is all that matters.

To vivify the contrast between theological doctrine and religious imagery, I begin with the classical problem of evil, a platform that prominently displays theological doctrine.

I. Anthropomorphism and the Problem of Evil

The argument from evil, so-called, purports to refute traditional theism. The fact of unjust suffering--some of it palpably so, grotesquely unfair--is, it is argued, logically incompatible with the traditional conception of God, specifically with God's moral perfection, omniscience, and omnipotence.

The argument relies upon what it takes to be the Judeo-Christian-Islamic conception of God. The conception to which the argument appeals is, so to speak, the received conception, within Judaism and without, certainly since medieval times. But the Bible and Rabbis, as noted, speak in quite a different idiom.

The project of distinguishing Rabbinic from medieval conceptions, however, is no trivial matter. Medieval modes of theological thinking are entrenched, by now second nature to us. Indeed, that there is any distance between the Biblical conception and definitions like Anselm's will come as a surprise to many. Furthermore, it is difficult to approach the Talmud except under the tutelage of our medieval teachers. It is especially difficult for Halacha, less so for Aggada. But even the latter is a substantial undertaking.

A. J. Heschel, an important American Jewish scholar and thinker, comments that much of his career was devoted to elucidating distinctively Jewish--as opposed to medieval Greek-inspired--modes of religious thought. "It is not an easy enterprise," Heschel notes.⁽⁹⁾

The Hellenization of Jewish theology actually goes back to Philo [first century BCE]… the impact of Philo on theology was radical. To oversimplify the matter, this approach would have Plato and Moses, for example, say the same thing, only Plato would say it in Greek…This view has had a great impact on much of Jewish medieval philosophy. They talk about God in the language of the Greeks.

One might say, mimicking Heschel's hyperbole, that Maimonides puts Aristotle in place of Philo's Plato and sees Moses as an Aristotelian.

Perhaps we should not suppose any failure on the part of the medievals to discern the distance between biblical-rabbinic and Greek inspired modes of theological thought. Rather for the medievals, this was the only way to make sense of the God of the tradition; the only way to square revealed and philosophic truth. Rabbinic tradition is speaking of this God, even if the Rabbis would not quite have seen it this way.

If this was the medieval tendency, Heschel's criticism may seem less damaging. One who wants to make philosophical sense of biblical/rabbinic remarks about God will inevitably do so in terms of one's own idiom and conceptual repertoire. It might seem, then, that the medievals only did what we all do, what is inevitable.

However, the interpreter of tradition needs to be alert to the possibility of gross imposition. As Heschel sees it, Philo's innovation violates something at the heart of the tradition. And it is not only a question of imposition. After such innovation, the tradition may take on new burdens, problems and puzzles that are artifacts of the new ideas. The classical problem of evil, I will argue, is an example.

The introduction of Greek modes of philosophical thought encouraged the minimization or outright rejection of biblical and rabbinic anthropomorphism. The God of the Hebrew Bible is, among other things, loving, nurturing/merciful,⁽¹⁰⁾ just, even angry. The Bible speaks of relations between people and God in the language of personal relationships. What grounds obligation to God, for example, is nothing very abstract. It is rather the community's historic and personal relation to God, a relationship that begins with the Exodus.⁽¹¹⁾

The aggadic imagery humanizes God even more. Just as we pray, God prays. And not only does he pray, but his prayers reflect, as it were, his worries--that in his treatment of his creatures his desire for strict justice will not overwhelm his nurturing, merciful side. Even more astounding, God prays wearing t'fillin, phylacteries, as do his people. Our t'fillin contain several Biblical passages, including the famous Sh'ma, "Hear, O Israel, Adonai is our God, Adonai is one [or is unified, or whole, or unique]." God's t'fillin contain a parallel passage, (roughly) "Who is like my people, Israel, a unique people?" God, we are told, weeps for his children, exiled from their homeland, their (and God's) Temple destroyed.

To take such imagery seriously is not to take it all literally. To talk of God's t'fillin, or of God weeping, is not to suggest that God has a body, an idea that does violence to mainstream Rabbinic understanding. But it is to think of God as the bearer of attitudes, thoughts, even feelings and vulnerabilities. The imagery of God's t'fillin suggests that God's stance toward his people is characterized by awe and love, the concomitants of our own recital of the Sh'ma, at least when we are sufficiently focused and prepared. But love involves vulnerability to the fortunes--even to the will--of one's beloved. Awe involves a certain humility.

Such robustly anthropomorphic characterizations of God play an apparently indispensable role in the religious life. That God loves and cares are, for the religious practitioner, no throwaways. To attempt to relegate anthropomorphism to the status of a mere surface level phenomenon is to engage in an uphill battle. The burden of proof, given the character of religious life, surely seems on the other side. On the face of it, the distance between biblical/rabbinic and medieval thinking is enormous.

Robust anthropomorphism presents insuperable difficulties for the medievals. A God whose love involves vulnerabilities is a God with limitations. We seem no longer to be speaking of than which none greater can be conceived.

Biblical/rabbinic anthropomorphism, then, seems to violate God's perfection. But it is the medievals' supremely perfect God that is pertinent to argument from evil. This suggests the problem of evil--at least as classically formulated--may indeed be an artifact of medieval innovation.

One may reinforce the point by reference to the perfections specifically relevant to the problem of evil. Heschel suggests that omnipotence--perfection in the realm of power--is not a biblical/rabbinic idea. He speculates that perhaps Islam, with its emphasis on the contrast between human subservience and the almighty God, bequeathed omnipotence to medieval religious philosophy. Whether or not he is correct about the history--his remark may be no more than his own Aggada-style comment--Heschel's reading of the biblical/rabbinic literature, his understanding of the Rabbis' stance on the question of power, yields a God whose awesome power is nevertheless limited.⁽¹²⁾

God's moral perfection is another good candidate for a second look. It is not obvious that the absolute moral perfection so important to the medievals corresponds or even coheres well with biblical/rabbinic tradition. It is at least an interesting question whether the God of the Bible and the Rabbis exhibits such ethical impeccability. Abraham, after all, argues with God that God's impending destruction of Sodom would be unjust; it would victimize the innocent along with the guilty.⁽¹³⁾ And then there is the rabbinic depiction of God's own prayer, that His desire for strict justice stays within its proper limits, that it stays in balance with His compassion. This sounds very much like a loving God who, aware of His limitations, is concerned that He, as it were, gets it right. And although the Rabbis do not emphasize the matter--as do the kabalists later--the God of the Hebrew Bible is a very complex deity, arguably one with a dark side.

To put the classical problem of evil to the side is not to deny that there are real issues in the neighborhood. The biblical/rabbinic literature does emphasize, after all, God's justice. What do we make of this? Furthermore, if there really is something to the idea of God's dark side, how does this cohere with His justice, indeed with God's being an object of worship? I hope to return to these very real problems of evil in future work. My topic here is the contrast between biblical/rabbinic and doctrinal ways of approaching religion.

II. Poetic Imagery and Religious Belief

1. Poetic Imagery

Let's consider a somewhat lengthier list of anthropomorphic images of God in biblical/rabbinic literature: loving and nurturing, even if demanding, parent; benevolent judge/ruler who does not forget acts of loving kindness and generously and lovingly passes on the rewards to one's progeny; righteous judge who has access to our deepest secrets and who rewards and punishes accordingly; king of the universe, to be treated with lordly deference; bridegroom; husband; woman in labor; angry, regretful, even vengeful, remembering the sins of the parents and visiting them upon even distant generations. When one scans this panoply of images, doctrine seems far away. Many of the images fail to yield anything like a characterization of God that could figure in doctrine. Among the images, moreover, are striking dissonances, hardly a doctrine-friendly phenomenon.

Clearly, conceptual refinement and coherence is not a high priority in the Bible and Aggada. One has the sense that one is dealing with something more like poetry; sometimes poetry per se, as in Psalms, other times poetic, image-laden prose.

The virtues of this poetry, any poetry, do not include the discursive articulation characteristic of philosophy. The sorts of things one seeks from poetry are brilliance and depth of perception, suggestiveness; these inextricably bound with beauty of formulation. A collection of poetry on the subject of, say, love might include pieces reflecting different attitudes, moods, experiences. The poet seeks to illuminate the phenomena, sometimes casting them in a positive light, sometimes in a negative light. Many of the images, perhaps even the most beautiful and suggestive ones, do not yield easily to anything like a philosophically adequate idea.⁽¹⁴⁾ Nor would the images presumably constitute a coherent set. Imagine the folly of trying to derive any sort of theory of love from such poetry.

Doctrine--the theory of God, as it were--is equally remote from biblical/rabbinic characterizations of God. The point holds not only for Psalms and the like; one of the fundamental ideas of Genesis, that humans were created in God's image, provides an example from Biblical prose. The idea of reflecting divinity is potent, pregnant with meaning. But it is imagery, not doctrine. Its very magnificence--literary and religious--seems to place it at some distance from doctrine.

2. The Language of Poetry

The central concept in the philosophy of language is arguably that of meaning or significance. What one takes meaning to be, to consist in, may well depend upon what form of discourse is in view. Starting from mathematics--and mathematical language was Frege's first love, or perhaps second after mathematical thought--or starting from scientific language as does Quine, or starting from poetry, or literature more generally, may yield very different ways of thinking about meaning. Analytic philosophy of language has come a long way in the past decades. But its practitioners have not explored--nor been interested in exploring--the question of whether or how its work might be extended to literary contexts.

The study of poetic language would require, no doubt, some re-tooling. Our hard won insights about meaning and reference may turn out to be relevant; it would seem unlikely for them to be altogether irrelevant. What is needed, though, is a fresh look followed by sustained attention. Here I can only make brief comments, barely a start.

One should not suppose that literary or poetic "imagery," any more than other forms of discourse, is necessarily connected with visual or mental images. Perhaps literary imagery is more likely than is say, a weather report, to stimulate one to form an image in one's mind. But that is not essential. We do not refer to these figures of speech as imagery because they induce mental images. How to characterize literary imagery--why exactly we call it "imagery"--is more difficult, something I won't explore here. But simple examples of metaphor, one form of such imagery, make it clear that mental imaging is not essential. "Sea of troubles," mentioned below, will do.

Here's The American Heritage Dictionary's definition of metaphor: A figure of speech in which a word or phrase that ordinarily designates one thing is used to designate another, thus making an implicit comparison, as in "a sea of troubles." When we explain metaphor to our students--say in connection with some philosophical metaphor--metaphorical language, we suggest, is second best. We revert to it when we are not in a position to provide literal description, when we can't do better than mere analogies, implicit comparisons. What we don't mention--because we are not really thinking about poetry and literary language--is that in literary contexts, metaphor is often used for its own sake; it is not second best.

Why is metaphorical language sometimes preferable? It goes to the heart of literary imagery--here we go beyond metaphor--that the words resonate. One might speak about this phenomenon in terms of levels or layers of meaning. But "resonance" better conveys the lingering intimations, the echoes, the movement of the mind.⁽¹⁵⁾ In poetry or poetic prose, a single word or phrase may have multiple resonances. Sometimes one predominates, comes to the fore; others linger in the background. Other times one finds oneself moving between them, sometimes repeatedly. Resonances have many determinants: the occurrence of the expression in famous literary contexts; or other contexts, or other ways, in which the words are customarily used; or other words that sound alike or that come from the same root--these are a sample.

Where does meaning, significance, come in? We should not try to say what the meaning of a word in poetry consists in--a bad idea even in, as it were, straight philosophy of language. Instead, let's say that an expression's significance has everything to do with the networks of resonance in which it is embedded. A host of just such miscellaneous considerations, for example, figures in the significance of "God," as well as of its many Hebrew correlates (or approximate correlates).

3. Mere Poetry?

Have I distanced the biblical/rabbinic literature too far from theological doctrine. If imagery is at the heart of biblical/rabbinic characterization of God, then what becomes of religious belief, of the tenets of religion?

Heschel writes "In Biblical language the religious man is not called "believer," as he is for example in Islam (mu'min), but yare hashem [one who stands in awe of the Lord]."⁽¹⁶⁾ In the present context this is extremely interesting; it suggests that in thinking about religion we make too much of the doxastic dimension and too little of the affective. What I take from this is not that the concept of religious belief has no purchase in Judaism. Or that the religious life does not require appropriate beliefs. Rather belief is not at the heart of the matter; one gets a misleading picture of Jewish religiosity if one's focus is a set of beliefs. There is an analogy with the question of linguistic meaning in poetry. It's not that a linguistic expression in poetry does not mean what it ordinarily does, say, in a newspaper article. But that dimension of its meaning is often not at the heart of its function in poetry.

What then do I make of religious belief, given my emphasis on poetic imagery? My answer is that likening biblical/rabbinic remarks to poetry certainly does not imply or even suggest that these remarks involve no beliefs, no real commitments; that they are, as one might say, "just poetry." For such commitment-neutrality is surely not true of poetry itself.

Poetry may assume, for example, straightforward factual information about the world, that there are people, that they behave in certain ways, and so on. Second and more interestingly, poetry may be committal even where there is no way to formulate the relevant belief in straightforward, literal language. If we wish to formulate such belief, we do one of two things. We can approximate, extracting a piece of the picture, one that is propositionally manageable, and attributing that piece. Alternatively, we can mimic the poet, attributing belief using the very imagery she used--or related imagery. This may be a philosophical no-no; it will strike some as bizarre to suppose that such a thing counts as belief. Here as elsewhere, as Wittgenstein urged, philosophy would do well to look at actual practice rather than think about what it must be like.

Before exploring this belief-reporting practice, let's turn back to the biblical/rabbinic literature. Here many things are assumed about the world, many beliefs can be distilled from the imagery. All the old standards, as it were: belief in God, in God creating the world, creating people in his image, freeing his people from Egyptian bondage, revealing himself to Moses and giving the Torah on Mt. Sinai, and the like.

How should we think about these beliefs? Are they like the straightforward propositional claims that we can often distill from poetry or are they of the second variety mentioned, claims that remain at the level of imagery? When we report someone as having such and such religious belief, to what extent are we discursively articulating propositional content; to what extent are we ourselves using religious imagery?

I'm not sure that the question can be answered across the board. Where the believer uses imagery, the reporter, using the same or closely related expressions, utilizes the imagery to characterize the belief. But even if one takes much of biblical/rabbinic talk of God's doings and ways to be impressionistic, is it all imagery? What of the term "God" itself, or more appropriately in the present context, the various Hebrew expressions for the deity? This leads to very difficult questions concerning those expressions. Do we really have a proper name for the deity? Do we have descriptions that apply uniquely? It is interesting--but of course only suggestive--that when Moses asks God his name, what he is given is something to think about. I'm almost inclined to say that in speaking of the divine, intimation is the rule. Perhaps attribution of religious belief remains at the level of imagery. Religious belief, one might then say, lives at the level of imagery. It goes without saying that the matter bears serious attention.

How then am I thinking about belief? What sort of belief is this that has no propositional content? In fact, on grounds that are completely independent of the current discussion, I think that the usual sort of thinking about belief--the propositional content model--is misconceived.⁽¹⁷⁾ Indeed, that the propositional content model is not consonant with my emphasis on religious imagery is fine with me. This is not the place to motivate my contrary conception. But we can at least see how it works out for the case of religious belief.

Let's distinguish two different ways of appreciating biblical/rabbinic statements about God, for example, that people reflect God's image (call it p, just to maintain respectability). One may appreciate p and its attendant imagery from the outside, as it were. Call this the Bible-as-literature approach. Alternatively, one may make p one's own, see the world through it, declare it, for example, in prayer. The question is whether one signs on, as it were; whether the poetic resonances reflect one's own take on the world.

If one wholeheartedly endorses p , then we may use p to characterize the person's commitments, to keep track of him with respect to how he approaches the world. He is, we might say, one of the p-endorsers. In so saying, we not only use p to keep track of him with respect to how he approaches the world, we use p to classify him with respect to others. It's like putting p at the top of the page and then listing p's endorsers.

This is what we do when we ascribe belief in p, when we say that he believes that p. Let p be as imagerial as you like, let it be an obvious non-conveyer of propositional content. If such imagery plays a fundamental role in one's approach to the world, then we can use p--its imagery intact--to characterize one.

To "read off" a belief from the imagery is thus to abstract a piece of the imagery, to kidnap it in a way, to absorb it into a different genre. It's to use the image in a new way. To ascribe such a belief to someone is like saying "This image plays a fundamental role for her."

Involved here is a sliding scale, from those parts of the imagery that are more serious and fundamental from those that are less so.⁽¹⁸⁾ Consider the imagery of God's right arm. It's as if one kidnaps the imagery of God's right arm, and then learns that one can't make straightforward use of it; it fails to go quietly into the belief ascription genre. This is so because it is an image we are happy to let go. At least we are after we learn from it. Religious Jews believe that God created the world, but not that he has a right arm. The latter is not part of the imagery with which they approach the world. It is not a constituent of the story that serves as a backdrop to their lives. So, as our belief reporting practices go, it is inappropriate to use this imagery to keep track of their religious whereabouts. This is not to say that the image has no power for them.

III. INTERLUDE: Is There Doctrine in Halacha?

Is it thinkable that even in the realm of Halacha the notion of doctrine has no application? This seems very unlikely. After all, the idea I have been advancing with respect to biblical/rabbinic characterizations of God, that conceptual refinement and coherence is not a high priority, is surely not true of Halacha. Talmudic study is notoriously wonderful training in analytical and conceptual skills important to philosophy. That this is so reflects the conceptual rigor implicit in the Talmudic texts, the emphasis on fine distinctions, on clarity. If the notion of doctrine has a home anywhere in Jewish thought, it certainly has a home in the realm of Halacha. But does it?

The tradition speaks of the sea of the Talmud; and when one includes the commentaries, one might well speak in oceanic terms. As one who has not ventured out far from shore, and in no more than a rowboat, I am not in a position to answer the question. But ideas are what our field is all about, and the anti-doctrine idea--even in the domain of Halacha--is at least interesting, perhaps even exciting. Or so I'll suggest.

When I was an undergraduate, being introduced to the joys of Talmudic study, the question arose about how Halacha is applied when social structures change and with the advent of new technologies. The Rabbis of the Talmud didn't know about electricity, for example, and so couldn't have prohibited its manipulation on the Sabbath. The explanation I received is that at the heart of Halacha are principles, and what happens over time is the application of these principles to new sorts of examples.

Leave aside the question of whether this is too simple, the sort of oversimplification one might supply to beginners. There is a certain picture here, even if it needs qualification and refinement. It is that underlying the particular prescriptions and proscriptions is a set of high-level principles, a system of axioms, as it were, from which the Rabbis draw inferences about actual cases, new situations, and the like.

Rabbi Soloveitchik, in Halakhic Man,⁽¹⁹⁾ likens Halacha to mathematics, and speaks of it as a set of a priori concepts through which "halakhic man looks at the world."

The foundation of foundations and the pillar of halakhic thought is not the practical ruling but the determination of the theoretical halakhah… practical decisions…do not stand at the center of [the] concerns of {many great Talmudists]….The theoretical Halakhah, not the practical decision, the ideal creation, not the empirical one, represent the longing of the halakhic man.

The resonances of both Plato and Kant are apparent here, and while the remark does not decisively endorse the high-level principle conception, it strongly suggests it. But consider the following from Nachmanides's commentary on Deuteronomy 6:18, "Do what is right (or "upright") and good in the sight of the Lord…."

…it is impossible to mention in the Torah all aspects of man's conduct with his neighbors and friends, and all his various transactions, and the ordinances of all societies and countries. But … [God] mentioned many of them --such as, Thou shalt not go up and down as a talebearer; Thou shalt not take vengeance, nor bear any grudge; neither shalt thou stand idly by the blood of thy neighbor; thou shalt not curse the deaf…and the like--[Therefore, since He could not mention all, God] reverted to state in a general way that, in all matters, one should do what is good and right…

Nachmanides makes a similar comment on the famous passage in Leviticus 19:2, "Be holy, for I, Adonai your God, am holy." He sees this passage as the same sort of general statement that goes beyond the particulars previously mentioned.

Nachmanides says that it is that it is impossible to enumerate any complete set of prescriptions and proscriptions. What we are given explicitly by God is rather an extensive set of examples. And then a "general statement." Together, these constitute enough to create a sense of what is required, a sense of what are natural extensions of the examples provided and also presumably of the limits, what would not be a natural extension of the examples.

The idea that the explicit commandments are only examples seems radical and quite exciting. But what is most crucial here is the character of what God states "in a general way," the character of this statement that goes beyond the explicitly mentioned examples. For these general statements are by no means high-level principles, discursive articulations of inclusive formulas, statements from which one can infer all the particular laws. They are more like the imagery that has been my focus in discussing Aggada. Indeed, they are just that sort of imagery. It's not, after all, as if a substantive account of "the good and the upright," or of "the holy" is provided.⁽²⁰⁾ The imagery of "the good and the upright" in its context, and of "the holy" in its context are sufficient, according to Nachmanides, to generate a sense of how to proceed, of what counts as a natural extension of the examples and what does not.

One way to conceive Talmudic legal debate and analysis--the arguing over and comparing cases, citing precedents, and so on--locates the real action at the level of underlying (or overarching) high-level legal principles. But there is an alternative. One might suppose instead that the real action is at the level of cases. Talmudic debate concerns natural and unnatural extensions, the sort of thing for which one can develop a feel in the absence of any grasp of high level principles. This second alternative recognizes, of course, that in the course of debate, there will be appeals to principles--legal, interpretive and so on. But the question is in the end, whether there is something like a system of doctrine--high-level inclusive formulae--that underlies the legal system.

There is an analogy here to the vexed question of our mastery of general terms in language. Is our command of general terms--like "intelligence" or even "game," Wittgenstein's favorite example--a consequence of our grasping general concepts, or equivalently, grasping general definitions? Or is the concept-talk a kind of mythology? Perhaps general principles are irrelevant; mastery is a matter of the learner's developing a feel for what are natural ways to extend the terminology from cases of known application.

Whether Nachmanides rejects the inclusive-theoretical-principle model for Halacha I do not know. But he does furnish an alternative model: The Torah provides examples and guides us in how they are to be extended. It's possible that he himself thinks quite differently about the character of Halacha at the deepest level. That is, although the Torah does not teach us the high-level principles explicitly--indeed it supplies only examples plus imagery--one can and should go on to formulate the relevant inclusive principles. But the thought that he might have rejected the very idea of inclusive principles is to me an exciting one.

Understood in this more radical way, Nachmanides would be playing Wittgenstein to Rabbi Soloveitchik's Plato. As Wittgenstein saw things, Plato is responsible for the original philosophic sin: the idea that what matters is not at the level of the particular example, but at the level of the higher, inclusive principle.

We have been considering the idea that the Torah teaches the law by examples supplemented by imagery. What is more certain is that the explicit prescriptions and proscriptions supplement the imagery, and in an essential way. For imagery, taken in isolation, is amorphous. Upright might get confused with overly rigorous. Intoxication with holiness might issue in a range of patterns of action, some at odds with how the tradition understands kedushah, holiness. So the proscriptions and prescriptions give content to the imagery. They guide its interpretation. This phenomenon seems quite general, applying to all the sorts of imagery I've discussed throughout this essay. That people reflect God's image, for example, could lead in multiple directions, and it is the ethical commandments that guide its understanding.

IV. Theology

1. Philosophical Ambition

The medievals, blessed with a more-or-less stable first philosophy, lived philosophically charmed lives. There is some parallel in our attitude to, say, physics. We are not sure of the details of our physics, not confident that we have final answers, but pretty sure we are in the ball-park, playing by the right rules, very confident of some general outlines. Perhaps the medievals were even more confident about philosophy. They were, after all, looking backward towards the Philosophers, while we are looking forward to a somewhat uncertain future, humbled by the history of past scientific upheavals.

Possession of philosophic truth grounded in a stable first philosophy makes many things possible: philosophy providing foundations for religious belief, philosophy clarifying the content of revealed religion, philosophy determining the nature of God, or perhaps determining that God's nature is demonstrably, as it were, beyond us. Doctrine, the propositional articulation of the religious fundamentals, is at home in such a setting.

A doctrine approach need not deny what I've been arguing about the impressionism of biblical/rabbinic characterization of God. One might argue--as Maimonides indeed suggests--that it is only because of philosophically available truths about God that we are in any position to discriminate among the images, to know which to take seriously and which not. The ordinary person has no independent access to God, so how can he discern Biblical talk of God's unity and God's creation of the world from the merely figurative attribution of bodily properties to God? Absent our access to philosophical truth and we would be religiously deprived.

This seems an extreme view, one that gives more independent weight to philosophy than many of the medievals, and probably more than Maimonides himself in other moods, or literary modes. But it nicely dramatizes the contrast with our own philosophic environment. The medieval conception (or cluster of them) of how religion and philosophy might join forces seems to many of us inapplicable nowadays. The crucial philosophic truth complement seems missing in action. It is not only that the traditional proofs of God's existence are in disrepute. Nor is it merely the lack of a received view--or even a widely accepted consensus--in philosophy, substantively or methodologically. For many of us philosophy simply cannot be brought to bear on religion he way that the medievals supposed.

One might assume that this makes it natural for us to think with the Rabbis, who also carry on without a first philosophy. However, the tradition, as we have it, has acquired much medieval philosophy, much doctrine, at least semi-officially. This of course was Heschel's gripe. If we follow his lead, trying to recover what he called Jewish as opposed to Greek ways of thinking about Judaism, we are indeed led straight back to biblical/rabbinic mode of religious expression.

To reject medieval philosophic ambition is not to suggest that philosophic training is irrelevant to the understanding of religious thought, Jewish or other. In philosophy we are trained to think carefully, analytically, to be sensitive to conceptual distinctions, to extend our thinking, as it were, both vertically, persistently pressing beneath the surface, and horizontally, taking a comprehensive view of the domain. There is no reason why such virtues would be irrelevant to the understanding of religion and religious ideas. But this is a far cry from the sort of access to philosophic truth that fuels the medieval project.

2. Theorizing about God

Woody Allen quips that many of the things our parents taught us were good for us are not so: milk, sun, red meat, college.⁽²¹⁾ I'd add doctrine to the list. If one tries to think with the Rabbis, it no longer seems a natural category. The Rabbis don't seem much interested in it. They don't seem driven by the quest for theorizing about God that Hellenistic culture bequeathed to Jewish tradition. With all their notorious questioning, with all the focus on clarity and fine distinctions in the Halachic realm, they fail to evince much interest in rigorous treatment of the theological. The nature of the being that lies behind the imagery is not a question that apparently kept them up at night.

Why is this? Why is theory not worthy of pursuit in this most important of realms? Even if one gives up on deriving the relevant theory from a first philosophy, or from philosophy at all, one might still attempt to come to intellectual terms with the theological. One might begin with the poetry and imagery of the Bible and Aggada. One might bring to bear insights gained from experience along with the analytical and dialectical skills exhibited by the Rabbis of the Talmud, virtues that seem closely connected with the virtues of philosophy, on the more modest conception mentioned at the end of the last section. Perhaps Halachic knowledge is also relevant. One cannot rule out the idea that knowing what God wants of us--the character of the religious life--is relevant. Reflection is reflexive; if people indeed mirror divinity, perhaps one can learn something about divinity by attending to its religiously developed human representation.

Even this sort of theorizing, however, seems not to have been on the rabbinic agenda. Why not? I do not know the answer to this question, but I do want to explore several sorts of considerations that may play a role.

a. What's It Your Business?

One idea, an obvious one, is suggested by reflection of what one might call Jewish sensibility. In Hannah and Her Sisters, Woody Allen's character is trying to explain his impending conversion to Catholicism to his parents. "If there is a God," he asks, "why was there a Holocaust?" His mother, rather hysterical, has locked herself in the bathroom. Her response to his query takes the form of an order to her husband: "Saul, you tell him!" Saul's answer is instructive, "Explain the holocaust; I can't even figure out how a can opener works." The sense that God's ways are imponderable runs deep in us. The possibility that by careful reflection we might come to an understanding of God's nature seems even more remote.

One might arrive at this last idea on the basis of serious theological inquiry; such a conclusion seems to be the upshot of Maimonides's thought about negative attribution. But the sense that runs so deep in Jewish culture is much less refined. It's the sense that the universe is a pretty crazy place, filled with wonders and with all manner of irony: the pleasures and comforts of family, community and God and his world, and the agonies imposed by those same agencies. The sense is not so theoretically grounded as it is in collective and individual experience. And one wonders whether this wasn't so for the Rabbis as well.

b. God's gift to us.

There is a second, related, sort of consideration that may shed light on the Rabbis' lack of interest in theology--in the sense of the theory of God. I have in mind a certain picture of what God has and has not given us, and of what we are to do with what we have been given.

What we have not been given is a theory of God. Nor does the Torah encourage us to suppose that we might attain to any such thing. The person closest to such things, after all, was Moses, and he never got a good look, even a straightforward statement of God's name.

What we have been given is a complicated business--there are many ways to come at it, and any particular way will be only partial. But somewhere near the heart of it is set of directions for how to live as Jews; indeed, how to live a life of holiness, both communally and individually.

This is not to say that there is in the offing anything like a theoretical treatment of holiness. The Rabbis, with holiness as with the theory of God, seem more tied to the earth. Their interest is not in a science of the subject, as it were. As a first approximation, we might say that their interest is in developing a manual, a set of instructions for a life of holiness. They were engineers, as it were, not physicists.

But even this is anachronistic, for a manual suggests a systematic arrangement, something like the Codes, the medieval codifications of Halacha. The Rabbis' interest, more accurately, is in a range that includes understanding, clarifying, sometimes refining the practices, both ritual and ethical. And we should not leave out their devotion to guarding the law, constructing fences around it, enacting further law that serves to maintain adequate distance from the original laws' violation. For convenience, I'll ignore the anachronism, and continue to speak of the manual.

At the heart of the manual is Halacha. One should not forget, however, the crucial role of its helpmate, the historical/dramatic materials of the Bible, supplemented by the Aggada. Without these, no set of do's and don'ts would suffice. Indeed, thinking of the practices as do's and don'ts is misleading, reductionistic. Both ritual and ethical practices, themselves, like poetry, resonate with meanings. Some of these meanings reflect the imagery, some the historic narrative, some the relation to other ritual and ethical practices. When we think of the giants of the tradition, models of the heights of spiritual development, their achievement was no simple matter of discharge of positive and negative duties. Their achievement reflects the way they approached those mitzvot (commandments), or even better, their approach--mitzvot in hand--to the universe and to life, their conception of how one is to carry on, with its behavioral, cognitive and affective components. So the non-Halachic materials constitute a crucial part of, or supplement to, the manual. And not to be neglected is role of exemplars, both the giants of the tradition and, closer to home, parents and teachers who model the form of life.

This, then, is what we have and have not been given; and what we are to do with what we have been given. One of the central practices is the study of the tradition, the most advanced form of which is the study of the Talmud. One learns a great deal about the practices by way of Talmudic study, but the value of the study is not thereby exhausted. Indeed, one studies tractates that, since the destruction of the Temple, have no practical application. So Talmud is a theoretical study. But there is theory and there is theory. In section III above, I raised the question of whether there is an overarching, high-level theory implicit in Halacha. The idea that I found intriguing is of a piece with the idea that in these other realms--the theory of God, the theory of holiness--the Rabbis were not theoretical minded. Might it be that the taste for theory in all of these domains is part and parcel of the Hellenistic legacy?

c. Poetic inconsistency

Let's return to our question: Why not start from the imagery and try to figure out what's going on with the leading figure? One of the aims of such theorizing would be to make sense of the tremendous multiplicity of roles that God plays in the imagery. What is this all about? What sort of being underlies these images and why are they fitting images for this being?

I begin with two examples that share some features with our situation. In both of these examples we begin with ideas that stand in conflict with one another. Think of this as analogous to the panoply of biblical/rabbinic characterizations of God, some of which are strikingly discordant with others. And in both examples something short of theoretical resolution is what seems natural and satisfying. Indeed, theoretical resolution seems out of place. These examples are mere analogies, but they will help illuminate one way of thinking about the theoretical reticence of the Rabbis.

There is a Hasidic adage that a person should carry in his pocket two pieces of paper with dissonant messages: "I am but dust and ashes," and "The world was created for me." I have often thought that the really difficult trick is to get both on the same piece of paper; to live a life that is not so compartmentalized, that integrates the superficially incompatible messages.

Maurice Friedman suggested to me that it would be better to leave them on separate sheets, maybe even in different pockets. The imagery of a single piece of paper suggested to Friedman what he took to be a bad idea: that the philosophic job is to render these insights coherent, to articulate an inclusive principle. What one needs is rather a kind of practical skill, the ability to negotiate experience respecting both truths--that is, both images, each of which illuminates human experience.

My second example concerns the universalistic and particularistic tendencies in Jewish religious thought. There are places in which the biblical/rabbinic literature reads as if what really matters is the human community. It is people, all people, who mirror divinity. Other times the virtually exclusive emphasis seems to be on our own community, the Jewish people. These are disparate ideas, and it is religiously crucial that we achieve a sense of balance, that we find coherence between them. Placing disproportionate weight on either represents a great loss. Overemphasizing the universalistic threatens the loss of the richness of the tradition; overemphasizing the particularistic threatens all manner of ugliness. Again, however, the sort of coherence we are after is not necessarily theoretical coherence in the form of an inclusive principle. More important is the ability to negotiate experience, appealing to one idea or the other when fitting, allowing each to call a halt when we are nearing excessive attention to the other.⁽²²⁾

Neither example is supposed to demonstrate the impossibility of producing a kind of super-principle, one that delineates the experiences in which one or the other of the opposing ideas is applicable. But--and this is more clear with regard to the universalism-particularism example--we don't even know the shape, so to speak, of such a principle. Do we, or ought we, have any confidence that the cases fall under some illuminating formula? Think about the cases in which one's universalistic scruples might usefully be brought into play. Ought we to be confident that all share some common feature or features, other than, of course, being cases in which particularism is getting undue weight? Even if one identified a plausible candidate principle, isn't it likely that it would have to be gerrymandered repeatedly to meet the needs of novel cases? While there remains the possibility of an inclusive principle, there is reason to be skeptical.⁽²³⁾

Both examples exhibit two features that are salient for us. First, there are conflicting ideas, each of which possesses a kind of truth or validity. Second, the tension is resolved not by discovering a higher order principle, but by the acquisition of a practical ability or skill, a kind of "knowing how." The agent develops a sense of balance, the ability to call upon the idea appropriate to the situation at hand.

How does the first feature apply to the sundry characterizations of God? Think of the diverse, sometimes conflicting, images of God as profiles, as views from a perspective. Each profile is crucial for the religious life. Each has validity. Each illuminates in its own way. There are situations in which the image of God as nurturing parent is salient. In other situations other imagery may be salient, perhaps God as an impartial judge, or as a friend, or as creator of heaven and earth, or as one you have wronged, or as the parent of one you have wronged.

There are still other situations in which two or more profiles of God are somehow salient. Some of these may be very pleasant; as if one were taking in several varieties of beauty at once, or through several sensory modalities. Some of these situations, though, may be troubling, confusing. Such situations, as Halbertal and Margalit point out in their excellent discussion in Idolatry,⁽²⁴⁾ are analogous to one who works for his father-in-law, who also happens to be his teacher, landlord, and plays unnamed other roles in his life. One can readily imagine situations that become quite complicated and confusing. One doesn't quite know where one stands.

Let's turn to the second highlighted feature: the resolution of the tension between the ideas takes place on the plane of action rather than theory; it involves a practical ability or skill. Theoretical resolution in the form of a comprehensive principle seems unnecessary. This is of course the critical issue in the present context. For theoretical resolution--at least a leading candidate for such a resolution--would involve an account of the entity that lies behind the profiles, an account of how these could possibly be perspectives on the same being.

How does this apply to the biblical/rabbinic imagery? The story here is more complicated than in the two examples. The religious life involves a combination of practical abilities or skills that are grounded in understanding, intuitive if not articulate. The religious life also involves habits, behavioral and affective. All of this requires education, training, and practice. As is our way with such things, some are more given to it, more gifted at it, some will take to it more easily than others, others may come along more slowly, but may attain greater heights in the end.

One stage in religious development is understanding the many different human relationships in terms of which these profiles of God are formulated: developing a sense of what it is to relate to another as child to parent, as subject to monarch, as defendant to judge, as creature to creator, as lover, as friend, and so on.⁽²⁵⁾ Developing a sense of these with their directions reversed is also of great utility: parent to child, monarch to subject, and so on. The more vivid one's grasp, the more deeply one sees into these relationships, the farther along one is in this stage of the training. Some of this understanding requires the accumulation--sometimes years--of experience. This education is a lifelong affair.

The next step--not that these need to be separated in time--is the application of this growing understanding to the relationships between people and God. One needs to think about and practice seeing oneself in relation to God as child to parent, with the variety of complications that entails; and to think about it from both sides of the relationship. And as lover to lover, friend to friend, judged to judge, and all the rest.

The payoff of one's work--the propriety and caring for others, the comfort and solace, the elevation and dignity that are the concomitants of developed religious character--depend upon one's ability to negotiate the world feeling and acting in ways appropriate to just such relationships. One needs to feel and act as if one has a Godly parent, a Godly lover or friend, a Godly judge who sees all, a creator of inexorable laws of nature that proceed as if we didn't exist,⁽²⁶⁾ even--I suspect--an angry, even vengeful Godly ruler--this last being more complicated and controversial.

Of course, one doesn't feel and act in these ways all the time, or all at the same time. Part of the skill--what takes training, practice, and experience--is to call upon, or be called upon by, the appropriate image at the appropriate time, sometimes a single image, sometimes multiple ones. The latter can be confusing, disconcerting, as Halbertal and Margalit point out, and it can be wonderful, sometimes both. At the death of a parent, for example, many of these images may strike: God as creator of inexorable laws of nature; God as friend and comforter; (and since belonging and community becomes so important at such times) God as focal point--glue, as it were--of Jewish community, a community that extends horizontally--the present community--and vertically--the community over time; and perhaps others.

Jewish ritualized prayer--something that also takes training and practice if it is to be more than mechanical (and even if it is merely mechanical)--provides another example of the sometimes confusing but wonderful multiplicity. In prayer, when it works, many of the magnificent images are summoned. One is provided with the opportunity of experiencing these relationships and of reflecting upon them, seeing more deeply into them, seeing new aspects all the time.

That there are multiple images, that they seem discordant--properties that make theory seem very far away--are thus rationalized. We don't do so by finding a theoretical account of God that puts the images in their right place. Rather the miscellany, the mixed multitude of robustly anthropomorphic ideas and images, facilitate the religious life.

In my discussion of religious belief (section II. 3.), I argued that we can make sense of that notion even within my anti-doctrine conception. At the same time, I evinced uneasiness with the notion of religious belief in the context of Judaism. The notion fails to get at the heart of the Jewish religious orientation. I want to connect that discussion with the present one, with the idea that the discordant images of God enrich religious life. Unfortunately my remarks here will be very sketchy; I'll return to the issue elsewhere.

As I explained my approach to religious belief, the imagery that plays a fundamental role for the believer finds its way into the report of the person's belief. Belief lives at the level of imagery. But the transition from the imagery as it functions for the religious person to the imagery as it functions in the report of her belief is a very significant one. For the original context is one of poetry, or image-laden prose. And in that context, consistency of imagery is hardly a virtue. Indeed, the very multiplicity and variety facilitate the religious life. But once the imagery has found its way into the context of belief, into a report of someone's belief, inconsistency becomes a substantial liability. This reflects the utility of belief talk, what it does for us.

The belief idiom is a fundamental tool for locating people cognitively, for tracking their cognitive whereabouts, their take on how things stand. For that very reason, it is quite sensitive to matters of consistency. To see this, don't think of beliefs about religious matters, but about the weather, or about history, or elementary particles. When, however, we speak of belief with respect to a domain in which imagery plays such a fundamental role, the idiom has limitations. Along with talk of belief comes the suggestion that it's not a good thing to have conflicting ones. And that simply is not true of the sort of domain in question.

Should we think of Judaism as a system of thought? What we have been given, thinks the religious Jew, are directions for living a life of holiness--as individuals but as ones whose flourishing demands community. The essential constituent of such a life is a system of communal and individual practice, both ritual and ethical, informed by a narrative history interwoven with religious imagery. The imagery, in all its variety and inconsistency, along with the historical narrative provide many dimensions of meaning to the practices. And the practices in turn give definition to the otherwise abstract and elusive imagery. The constellation of practices, historical narrative, and imagery issue in a distinctive kind of life with its own substantial virtues and rewards. It seems reductive and misleading to represent this as a system of doctrine, a set of well-formed beliefs, a system of thought.⁽²⁷⁾

ENDNOTES

1. I owe this way of formulating the tendency to Joel Gereboff, who is also skeptical about the supposition.

2. When I speak in this essay of the Bible, I mean to refer to the Hebrew Bible. This is not quite the Old Testament of the Christian Bible, both in terms of the order of the books--the prophets come immediately after the Pentateuch in the Hebrew Bible--and the addition in the Old Testament of certain books not in the Hebrew canon, e.g. Maccabees.

3. Roughly, first to fifth century of the Common Era.

4. The role of anthropomorphism in the contrast between the two traditions constitutes a central focus of Idolatry, M. Halbertal and A. Margalit (Cambridge, Mass. and London, 1996).

5. I will use "doctrine" to refer to the output of philosophical theology. While I want to deny the appropriateness of doctrine, I certainly do not deny the appropriateness of religious belief or religious tenets. More on the distinction later.

6. Not that the Rabbis always speak in quite the same ways as the Bible. See Halbertal and Margalit, p. 31 ff.

7. Perhaps it would be better to say that the Kabbalists provide a more transcendent characterization of people and of the world.

8. There are relatively minor differences in practice that derive from different historic Jewish communities, for example, Sephardic as opposed to Ashkenazic, as well as differences that reflect different local customs or the different rulings of different Rabbis on questions of law.

9.

10. The Hebrew word rachamim, often translated "mercy," might better be seen as conveying nurture or something with both dimensions. Mercy has, for me at least, connotes grace, and this seems not to capture the feel of rachamim. The Hebrew word may be etymologically related to the word for womb, rechem.

11. I'm indebted to Halbertal and Margalit for this point about the basis of obligation being personal. See p. 31.

12. See "Jewish Theology," pp. 154-163, esp. p. 159 in Moral Grandeur and Spiritual Audacity, Abraham Joshua Heschel, ed. Susannah Heschel (New York, 1996). Heschel notes on p. 160 that he has "many serious reservations" about the Maimonidean theological system.

13. There are ways of reading such biblical passages according to which they do not suggest any divine limitation. Still, God's ethical impeccability is at least an interesting question in connection with such passages.

14. Such imagery often simply is not subject to explication in propositional terms without remainder, as it were. For one thing, many poetic images are propositionally indeterminate; were one to work at explicating or articulating the imagery long enough, one might see several ways it might be explicated, none of which is dictated or even suggested by the original. This is a bit like what Carnap thought about the explication of terms of ordinary language. But analogies with Carnap's views aside, the point here is that imagery is what it is; it does not yield to propositional formulation.

15. > The superiority of "resonance" over "levels of meaning" was suggested to me by Rachel Adler.

16. In God in Search of Man" (New York and Philadelphia, 1959) , p. 77, See also Richard Friedman, in The Disappearance of God (Boston 1995):