Conservatism and Innovation

1996

A.

When the Little Prince reached the lamplighter's planet[1], the Prince greeted him with due respect: "Good morning. Why have you just put out your lamp?" "That is the order," answered the lamplighter. "Good morning." "What order?" "To put out my lamp. Good evening." And he lit it again. "But why did you light it again?" "That is the order," answered the lamplighter. "I do not understand," said the Little Prince. "There is nothing to understand here," said the lamplighter, "an order is an order. Good morning." And he put out his lamp. Then he wiped his brow with a red checkered handkerchief. "I have a terrible job. It used to be a reasonable job. In the morning I would put it out and in the evening I would light it. The rest of the day I could rest, and the rest of the night I could sleep…" "And since then the order has changed?" "The order has not changed," said the lamplighter, "and that is precisely the trouble! Every year the planet turns on its axis more quickly, and the order has not changed!" "And then?" "And now, since it makes a full turn in a minute, I do not have even a single second of rest. Every minute, I light and put out!" "That really is funny! A whole day where you live lasts a single minute!" "That is not funny at all," said the lamplighter. "We have already been standing here talking for a month." "A month?" "Yes. Thirty minutes. Thirty days! Good evening." And again he lit the lamp. The Little Prince watched him, and, of all people, this lamplighter, who carried out orders with such faithfulness, pleased him. He remembered the sunsets he himself had wanted to see on his own planet, and for that purpose he would move his chair each time. And he wanted to help his friend: "You know… I can tell you how you can rest whenever you want to…" "I always want to," said the lamplighter. For a man can be both faithful and lazy. And the Little Prince added: "Your planet is so small that in three strides you can walk around it. You need only walk slowly enough to remain always in the sunlight. When you want to rest, simply walk… and the day will last as long as you like." "That does not help me very much," said the lamplighter, "what I love in life is sleep." "Too bad," said the Little Prince. "Too bad," said the lamplighter, "good morning." And he put out the lamp. "That man, all the others would despise him," the Little Prince said to himself as he continued on his journey: "the king, the conceited man, the drunkard, the businessman. And yet, to my mind, he is the only one who is not ridiculous. Perhaps because he is not occupied with himself, but with something else." And he sighed sorrowfully and reflected further: "He is the only one with whom I might have become friends. But his planet really is too small. There is no room on it for two…"

This long quotation, from the well-known children's book, presents with astonishing clarity many characteristics of the conservative type:

1. He is unwilling to depart from the instructions given to him even if the conditions in which he acts have changed (the planet is spinning faster).
2. He is diligent and very exact in carrying out those instructions.
3. This diligence does not contradict a lazy disposition, and at times it even stems precisely from it (an unwillingness to examine the fundamental assumptions).
4. Many of the others despise him.
5. There is great room specifically to appreciate him, contrary to the usual attitude toward him, precisely because he is not occupied only with himself, unlike all the 'rational' ones (the king, the conceited man, the drunkard, and the businessman).

And finally,

6. On his planet there is no room for two; it is too small. The conservative does not accept approaches other than his own.

The lamplighter is a conservative, whom the Little Prince tries to persuade to reconsider his mode of action in light of changed circumstances. The problem of conservatism versus innovation accompanies almost every ideological/religious movement as it attempts to cope with changing circumstances: Zionism, Judaism (as religion and as nation), Christianity, communism, and so on. At present, the debate in our own setting between Orthodox Jews and Reform, Conservative, and even secular Jews has become more topical, over the question of who represents the continuation of the Jewish tradition. Who is changing and who is consistent? Claims such as 'Jewish law has always adapted itself to reality,' and therefore, 'whoever advocates the approach that Jewish law must not be changed is the true innovator,' are heard constantly. These claims are also intertwined with arguments about the genuine irrelevance of parts of Jewish law, although ostensibly they belong to two different categories of attitude toward Jewish law. A public discussion of this kind, by its very nature, does not deal with the real problems and real differences between the approaches. Before one forms a worldview regarding such questions, it is necessary to clarify as thoroughly as possible the basic terms used in the discussion, apart from the concrete problem under consideration.

In the following pages I will try to characterize and analyze the various attitudes toward systems of laws/rules, ideological or otherwise, and I will examine the assumptions underlying the basic approaches: conservatism versus innovation or change. In my opinion, this is the first step in clarifying questions such as that of the attitude toward Jewish tradition. The next step in such clarification should be an empirical examination of the history of Jewish law, and an application of the abstract categories presented in these pages. That step, of course, exceeds the bounds of a priori philosophical discussion, and therefore has no place here. The relevance of the present discussion to current problems serves me here only as motivation for undertaking it. This article does not deal at all with the specific example of the system of laws of Jewish law.

To conclude the introduction, I will add one more important remark. In the discussion below I will occasionally use expressions that may sound like personal characterizations of the types themselves (the conservative and the innovator), but that is not the case. I have no intention at all of dealing with psychological layers of conservative or innovative behavior, but only with the philosophical characterizations of these modes of thought and attitude.

B.

I shall begin the discussion with a simple model. A group of people has been given a list of instructions according to which they and their descendants are to act throughout their lives. During the years of activity, the circumstances in which the members of this group act change, and as a result two subgroups develop, representing different approaches to the set of instructions: the conservatives and the innovators. I will now try to characterize the difference between them. The first formulation that comes to mind is:

Formulation A: The conservatives are those who continue to act according to the previous instructions even under changed circumstances, whereas the innovators adapt the instructions to the new circumstances.

Let us take an example that will illustrate the above situation.[2] Two people are walking in the desert dressed in swimsuits, just as their ancestors have done from time immemorial. After many years, they arrive in a very cold region. Person A continues to walk in a swimsuit, whereas Person B changes his clothing to warm clothes in accordance with the new weather conditions. At first glance, the classification according to the above characteristics seems simple and clear: A is a conservative and B is an innovator. But a somewhat more careful examination of the situation shows that matters are not so simple. A's claim is that he continues the way of his ancestors regardless of changing circumstances: just as they wore swimsuits, so does he. But B can claim, against him, that specifically he is the conservative, for just as his ancestors wore clothing suited to the climate in which they acted, so does he: he wears warm clothing in a cold place.

And the Hasidic story is well known about R. Noah of Malchovitz, who took over the leadership in place of his father, R. Mordechai, and whose disciples saw that in certain matters he did not conduct himself as his father had. When they asked him about this, he replied: "I conduct myself exactly like my father. Just as he did not imitate others, so do I."[3]

This dilemma calls to mind Wittgenstein's discussion of following a rule. In that discussion Wittgenstein asks[4]: Someone who promises, day after day, "Tomorrow I shall come to visit you" – does he say the same thing every day, or each day something different?

If we return to our acquaintances walking about in swimsuits, and to the dispute raging between them over who deserves to be called a conservative, it seems to me that the natural inclination is nevertheless to say that specifically A is the conservative. If I try to translate this classification into a more general characteristic, a formulation different from the previous one emerges for the difference between these two types:

Formulation B: The conservative is the one who interprets the instructions according to their plain meaning, whereas the reformist allows himself interpretive freedom. This interpretation is supported by thinking about the logic or the values underlying the instructions.

This formulation recalls Wittgenstein's words in the same discussion:[5] We merely read the lips of the rule and act, and do not turn elsewhere for any further guidance. That is, the conservative is the one who really acts according to the rule, and not the one who requires an additional interpretation, which can certainly be attributed to his own system of thought rather than to the framer of the rule. An action of the reformist type, according to Wittgenstein, cannot at all be defined as action according to that rule.[6]

Formulation B apparently assumes that the pair mentioned above possesses a text (written or oral) specifying the instructions, and in this case: 'one must walk about dressed in swimsuits.' If so, it is clear that A's interpretation is the one better suited to the plain meaning of the text of the instructions.

If the pair has no such text in its possession, but only the custom of their ancestors, then the decision between the two claimants to the crown of conservatism depends on the interpretation of the ancestors' practice. A argues that their ancestors acted thus because they were instructed always to wear a swimsuit, whereas B argues that the instruction was to wear clothing suited to the weather. In such a case, apparently, one cannot use Formulation B in order to characterize the two types, since there is no written wording at all, and no possibility of giving it an interpretation. The practice of the ancestors, however, can be interpreted in both ways. In such a case, if indeed the original wording of the instructions was as B interprets it, it seems that specifically he is the conservative.

Yet here there is room to argue that even if the instructions were to wear clothing suited to the weather, there is no justification for saying that A is an innovator. For in the formulation above, the innovator gives a different interpretation to the instructions on the basis of the logic underlying them. It is very difficult to say that A gives an interpretation to instructions that state: 'one must wear clothing suited to the weather,' and understands the value underlying them to be that one must wear a swimsuit in every climate, for there is no logic at all in such an interpretation.

Here it seems that the characteristic of conservatism or innovation depends on the interpretation of the data possessed by the group, and not on the original instructions themselves. If the data are only the facts that the ancestors walked in swimsuits, then the conservative will infer that the instructions were probably to wear a swimsuit at all times, whereas the innovator will argue that the instructions were to wear clothing suited to the conditions. From the analysis above it emerges that, according to B's approach, which understands the instructions as the statement 'one must wear clothing suited to the weather,' there is no fitting classification for A's approach at all. He is not a conservative, of course, but neither is he an innovator (at least according to the definitions in Formulation B). By contrast, if the guidelines are as A understands them, namely: 'one must wear a swimsuit,' then B's classification as a modifier is reasonable. It seems that in such a situation the reasonable person will say that A appears to be the conservative and B the innovator. This determination is merely an instance of the interpretive principle known as the principle of charity, since the opposite determination would complicate our attempt to understand and define A.

The above classification, which relies on the principle of charity, seems misleading. One cannot classify B as an innovator merely because, if he were classified as a conservative, we would have no fitting classification for A at all. The correct distinction here is that these two have no common measure (incommensurable), and each must be evaluated within his own interpretive context. There is no necessity that every interpretive context allow a rational characterization of all the other approaches. This point raises the problem that even the determination of which of the figures above is conservative and which innovative is itself dependent on interpretation. If so, the categorical system that determines the matter is that of the interpreter who looks at them from the side and tries to characterize them.

If we nevertheless try to discuss the problem objectively,[7] it seems that almost every interpreter will say that both these types are conservatives, each in his own way. Such an interpretation departs from Wittgenstein's strict requirements for one who acts according to a rule, as they were interpreted above. Here we shall say that although there is a dispute between them as to the nature of the instructions according to which they should act, whether they are constitutive or purposive – that is, whether the instruction is the concrete practical command, or the value underlying it – still, subject to each one's assumption that he is correct in his interpretation, both adhere to the instructions they received from their ancestors and strive to live by them.

Even in the first case, where the group possesses a wording of the instructions, it is still difficult to justify why one interpretation is more faithful to it than the other. And if the faithfulness of the two interpretations is similar, then again there is discomfort in defining A as conservative and B specifically as the one who changes things.

The unavoidable conclusion is that the two types we have presented thus far are both conservatives: one is a constitutive conservative, who understands the rule in his possession as constitutive, and the other, who understands the rules as directed toward the values underlying them, is a purposive conservative.

This is the place to note that I do not intend to claim that the constitutive conservative acts irrationally because he carries out the instructions without understanding the logic or the values that underlie them. Even A, who apparently acts without any connection to the logic underlying the instructions, is usually motivated by some rational motive. That motive may be trust in the giver of the instructions, and the belief that there is a logic in the instructions even if he does not fully grasp it[8]. In any case, a clear characteristic of the conservative is that he does not understand in full the logic underlying the instructions, for otherwise he has no reason at all not to depart from them when circumstances require it[9].

It seems, then, that we still have not identified the true reformist. At first glance, we are left now only with the possibility of defining the true reformist as one who denies altogether the ultimate obligation to obey the instructions received by previous generations. If so, we may formulate the matter as follows:

Formulation C: The conservative (of whatever kind) is one who aspires to obey the instructions as his ancestors received them (according to his interpretation), whereas the reformist is one who does not acknowledge at all any obligation to heed those instructions.

It is immediately apparent that this formulation too is deficient. The reformist too has some connection to the instructions, and does not recommend abandoning them entirely. For were that not so, he would simply leave the group that obeys them, rather than recommend reforms in the system of instructions. The type described in the last formulation as an innovator is better called a rejector.

To sum up, we have defined three groups with different modes of relation to the system of instructions: the group with the constitutive conservative attitude, which preserves the rules according to their plain meaning; the purposive conservative group, which preserves the rules while giving them a non-ossified value interpretation; and the group of rejectors, which is not interested at all in preserving the instructions.

The problem that now remains before us is that this theoretical division into three groups does not exhaust all the types familiar to us from reality. There is a reformist attitude that we have still not defined. The reformist type wants, in certain senses, to preserve a connection to the old instructions, but takes for itself the right and freedom not to preserve them in full. Characterizing this group seems rather difficult. The threefold division presented above seems to cover the entire range of possible interpretations.

It nevertheless seems possible to raise another formulation of the difference between the reformist and the conservative:

Formulation D: The reformist is conservative with regard to some of the instructions and a rejector with regard to others, whereas the conservative relates in a similar way to the entire system of instructions.

This still does not seem to be an exhaustive definition of this type, since with regard to a system containing only one isolated instruction, no such approach is possible at all. Or, in another formulation, with regard to each isolated instruction within the system, the reformist is either a conservative or a rejector. In addition, according to this proposal one must specify the nature of the revisions that this type can propose. Are proposals to add instructions possible within such an approach? If so, what will be the status of these additional instructions? Will the attitude toward them be conservative or rejecting? Are proposals to remove instructions or alter them[10] what characterizes reform? And if so, instructions of what kind? Altering instructions of the type toward which this model of the reformist is conservative is not a live option. As for those toward which he is a rejector, he has no reason to remove them (just as the rejector himself has no such reason). Another possibility that should be examined is that removing precisely those instructions he rejects – that is, cleansing the system of the irrelevant instructions – is itself the essence of reform. This is in contrast to the full rejector, who has no reason at all to deal with the system of instructions, since he does not believe in it at all.

According to this possibility, the reformist is a partial conservative. That is, he is a complete conservative, except that the system of instructions he preserves is different from that of the earlier conservative. This does not sound like an essential difference in attitude. In addition, there is room to examine the source and force of the instructions, and how it is possible to erase some of them while leaving others. That is, generally speaking a system of instructions obligates as a whole; otherwise it would not be a system of instructions but a collection of isolated instructions. With respect to each instruction on its own, there is the above argument, which allows no third category. It is clear, then, that the present definition of the reformist requires a web of instructions connected to one another. Such a picture does not allow the deletion of parts of it by one who accepts it as a whole. For such a system is logically equivalent to a system with one instruction, regarding which we have already shown that only two approaches are possible: conservatism (of its various kinds) and rejection. And if we say that one who erases parts of the system does not understand it as an inseparable web, then this is again merely an incidental difference in interpretation of the system of instructions, and not a real difference in attitude toward it. We should also note that even with regard to what remains of the system of instructions, obedience to it stems from the reformist's decision, and not from his full subordination to the instructions.[11] Such an approach hints that even the reformist's conservatism is a conservatism difficult to define in our previous terms.

 

C.

In order to understand the source of the attitudes described in the previous section, let us try to examine, in the examples brought above, what assumptions could underlie them.

When the walkers arrive in the cold region, B can justify changing his clothes by saying that this is in fact what the giver of the instructions intended, or alternatively by saying that there is no need at all to obey him. As long as we walked in the desert there was no reason to depart from these instructions, because they accorded with common sense; but now, when they cause discomfort, one can simply throw off their yoke. The prince in the passage quoted above did not intend to claim that there is no need to light lamps at night, but only that this action requires a great deal of effort in the present situation (when the planet revolves at a much faster rate). An examination of the roots of such an approach may show that in fact this is the same approach as that of the rejector, which assumes at its base that from the outset there was no reason to act according to the instructions, and indeed we did not act according to them. Walking in a swimsuit or lighting the lamps was our choice and not obedience to any external authority whatsoever.[12]

On the other hand, it may be that this is indeed a new approach, which claims that obedience has considerable value but that it is not ultimate. Obedience still requires examination as to whether its benefit outweighs its cost. The cost as well is measured within some system that is also accepted by that person. That is, the evaluation of obedience to the instructions is no longer carried out within a framework of binary logic (one must obey or one need not obey – conservatism or rejection), but within a framework of multi-valued logic, or fuzzy logic. If so, we have finally arrived at a plausible definition of the difference between the conservative and the reformist:

Formulation E: The conservative (of its various kinds) is one who relates to the system of instructions in an ultimate way, whereas the reformist is one who gives this system a value that is not absolute.

According to this formulation, contrary to our previous argument, one can imagine a reformist approach even to a system of instructions that contains only a single instruction. For example, the instruction to light lamps only at night can be received in one of four modes of attitude: absolute obedience – constitutive conservative; obedience according to relevant circumstances – purposive conservative; obedience on condition that the price is not too high – reformist; and disobedience – rejector.

The lamplighter cannot indeed be classified as a reformist, and certainly not as a rejector, but he can definitely also be called a purposive conservative. That is, the cost of fulfilling the instruction (the effort) is not at all relevant from his point of view, but it may be that if at some point there were no night at all on his planet, he too would cease engaging in the fulfillment of this instruction, and would understand it as irrelevant.

In light of all that has been said above in defining the reformist approach, let us now try to wonder about the source of conservatism itself. We generally tend to think that the constitutive conservative does not understand the full logic standing at the basis of the system of instructions in his possession. By contrast, the purposive conservative thinks that he truly understands it properly, and therefore allows himself to interpret the instructions in accordance with changing situations and in light of the logic underlying them. Unlike the reformist, both kinds of conservatives assign ultimate value to the system of instructions in their possession, and therefore are not inclined to replace it. It would seem, at first glance (and especially in the religious context), that they are motivated by belief in the capacity and intentions of the giver of the instructions. For that reason, the constitutive conservative carries out these instructions even without understanding their value. The purposive conservative thinks that he understands the intention of the giver of the instructions, and therefore allows himself greater interpretive freedom, but he too does not depart from them at all.

However, one can also present a different conception of the conservative. This type thinks that the system of instructions in his possession is a basic value system that needs no justification and is not susceptible to justification. On the contrary, this system is itself the evaluative standard by which all actions and situations are measured. In Kantian terminology, this is the person's transcendental system.[13] This is a somewhat surprising characterization of the conservative. According to this characterization, there is no need for a giver of the instructions and for full trust toward him, as the conservative was presented in the previous conception. This change in the definition of the conservative is something like Kant's Copernican revolution.

If we examine according to what criteria the rejector rejects the system of instructions, or alternatively according to what principles the principles are interpreted according to the purposive conservative, we see that they use a prior value system in whose light this system is interpreted or judged. From this argument it follows, in fact, that these too are constitutive conservatives, except that they use a different constitutive system. This system is not the system of instructions under discussion, but another one that they regard as prior to it. For them, the constitutive system is the value system according to which they judge even the present system.

When a person acts within such a basic system, there is a tendency to interpret other systems by reduction to the constitutive system. For example, one who believes in the validity of natural morality will often tend to interpret the laws of religion in a way that accords with the rules of natural morality. It follows, then, that one who has a clear constitutive system, yet at the same time feels obligated also to an additional system, will tend toward a purposive interpretation of the less basic system, in such a way that acting according to it will not contradict the rules of the basic system.

One may therefore think that every person is a constitutive conservative, except that his constitutive system is not identical to that of other people. Some of those constitutive conservatives will tend toward purposive interpretation of less basic systems, and will reject systems that are not open to such reduction (they will deny them).[14] In such a case it is clear that the discussion about whether a person is conservative (of whatever kind) or a rejector must be conducted on the plane of a given system of instructions. In practice, the discussion concerns the question: what is the place of this system in relation to other principles within the total thought-world of the person under discussion? If it is the most basic, then we determine that this person is a constitutive conservative with respect to this system; and if it is not such, then the way is open before him to be a purposive conservative. If this system is altogether absent from his thought-world, then he is a rejector.

The conclusion of the foregoing is that the discussion of conservatism is not, as it appears at first glance, a characterization of a person, but a characterization of his relation to the system of instructions. It would seem, therefore, that one can reduce the discussion of a person's conservatism to a discussion of whether the system under consideration is constitutive, purposive, or irrelevant for him.

Despite all that has been said above, it is customary to describe some people as more conservative than others. The term conservative is considered as describing people, and not only systems of instructions. As I noted at the end of the introduction, I do not intend to enter into a description of the conservative's overall character, but only to say that the concept of conservatism generally appears as a characteristic of a person's relation to a system. By contrast, the discussion of whether a system of instructions is constitutive or purposive is perceived more as a discussion of the system as such. It is true that even there it is difficult to detach the discussion from the intentions of the framer/giver of the instructions and of the one who acts according to them, since in the final analysis they are the ones who determine the validity and meaning of the system of instructions. The degree of a person's conservatism with respect to a particular system is in fact a description of its place within his overall value system. From this distinction it is clear why the descriptions purposive conservative and constitutive conservative have parallel descriptions in the classification of systems of laws as constitutive or purposive, whereas the description of the person as a rejector or as a reformist has no such parallel at all. Describing a system of instructions as irrelevant or as non-absolute is not a description of the system, but of the person's relation to it.

To sum up, there are two kinds of systems of laws: constitutive and purposive. There are four kinds of attitudes toward systems of instructions: constitutive conservative, purposive conservative, reformist, and rejecting. These labels characterize a person's relation to a particular system, and not an overall characterization of his personality. The discussion of whether someone belongs to one of these groups is not detached from the discussion of the character of the system itself, but the subject of the discussion in this case is the person himself and his relation to the system of instructions. That relation is derived from the place of this system within the framework of the person's overall world of values.

D.

From the analysis in the previous section several pointed questions arise. The place of the various systems within the overall value framework is conceived as something that does not lie in the person's decision, but as a datum prior to all his judgments. The judgments themselves are merely applications, to particular cases, of principles already embedded in advance. Does such a conception accord with approaches that are not deterministic? Another question – and perhaps this is another aspect of the same question – is how the reformist conception fits into this whole structure. Is a reformist relation (that is, a contingent one, according to the definition above) to some system of instructions possible only when that system is not relevant from the standpoint of the person's entire given hierarchy of values? And further: are a person's value judgments really made in the form of a hierarchical axiomatic system as described here? Is it possible in some way to change the system ingrained in the person itself? How, according to such a description, is it possible for a person's attitude to change from a conservative relation to a rejecting or reformist one? If such a transition occurs, as we are usually inclined to say, was the previous classification of this person's relation to the system of instructions an illusion?

In the structure described above, the conservative relation to a system of instructions is presented as the placing of that system in the highest position on the person's scale of values. A conservative relation is in fact the conception of that system of instructions as transcendental. It is patently clear that what is involved here is a subjective transcendental system, varying from one person to another. Such a conception runs counter to Kant, and to various transcendental arguments that came in his wake, which of course take for granted the unity of transcendental concepts and rules.[15]

In common arguments in ethics and aesthetics, the concept of 'value' is posited as a transcendental concept. Yeshayahu Leibowitz, who is himself known for holding such an ethical conception, brings two quotations representing such a stance, one from Pascal and the other from Maimonides.[16] Usually (as also in Leibowitz there) the argument is presented as arising from the agreement of all humanity, and therefore this is not the transcendental pluralism whose existence I claim here. But I do not see why thinkers who advocate ethical relativism could not also agree to such a transcendental argument, even if the result is not necessarily uniformity of values.

The source and validity of a value system is a topic that many have discussed, and this is not the place for it. We shall remain here at the analytical level of the discussion, without drifting into metaphysics, epistemology, and perhaps even psychology, as would be required for dealing with the above questions. At such a level, one may say that an argument that in this context can be described as a second-order transcendental argument should lead us to the conclusion that a person has an autonomous capacity to organize the hierarchy of his systems of values/instructions freely. This process of organization is not carried out in an axiomatic fashion on the basis of a prior value system, for otherwise the question would recur as to how the superior place of that prior system within the hierarchy was itself determined. This point of free will is also what makes it possible for him to relate contingently to some of the laws and to be a reformist or a rejector. The claim that this process itself is not carried out through a system of laws also removes the basis for the question about the possibility of changing the basic system. One may say without contradiction that such a change is possible, if it too is carried out in a non-algorithmic fashion, like the initial arrangement.[17] My purpose in this article was only to define the different attitudes toward systems of rules, and not to investigate where they come from. Dealing with these questions involves recourse to fields such as those noted above, and it greatly exceeds the framework of this article.

[1] The Little Prince, Antoine de Saint-Exupéry, Am Oved, translated by Ilana Hammerman, 1995

[2] My thanks to Amnon Levav of Jerusalem, who raised this example in one of our many discussions.

[3] See, for example, in the anthology 'From the Treasury of Hasidism,' from 'Or HaGanuz' by Martin Buber, the Culture Section of the Department of Education and Culture in the Diaspora of the World Zionist Organization, Jerusalem, 1970, p. 11.

[4] Ludwig Wittgenstein, Philosophical Investigations, translated by Edna Ullmann-Margalit, Magnes Press, 1995. See there sections 185-242. The example is taken from section 226.

[5] Wittgenstein, section 228 there.

[6] There is certainly room to discuss further the relation between this formulation and Wittgenstein's own formulation, and one may also disagree with the interpretation I have offered in the body of the text. It is not my intention here to clarify his doctrine, and for the clarification of conservatism and innovation, what has been written here is sufficient for my purposes.

[7] 'Objective' here means detached from the conceptions of the two sides being characterized.

[8] The claim of action with no motive at all does not appear plausible even when describing A's mode of action (the constitutive conservative).

[9] There may be a pathological case in which the conservative understands the logic in the instructions, but in every case fails to see circumstances that contradict that logic. In such a case the conservative is in fact a degenerate case of an innovator, who likewise, when he does not encounter circumstances that require change, is not inclined to amend the instructions.

[10] In the simple conception, changing an instruction is not a basic act. Such an act is equivalent to removing one instruction and adding another in its place. Therefore the arguments against change are composed of arguments against removal and addition.

[11] For this reason, in the Jewish tradition, anyone who deletes even one letter from the Torah is a heretic, that is, a rejector. The reason is that the remaining letters he leaves intact already have the significance of autonomous activity and not of obedience to the set of instructions. Whoever has it in his power to delete one instruction clearly assumes that he has it in his power to decide, with respect to all the instructions, which to reject and which to retain.

[12] For example, one who claims that it is worthwhile to continue walking in a swimsuit in order to preserve uniformity of appearance within the group is clearly, in basic type, a rejector, except that the adoption of the system of instructions serves his own purposes, and therefore he recommends that we continue to 'obey' them. A good example of this argument is Ahad Ha'am's conception of religion as the national 'glue.' The classification of concrete types and their comparison to the theoretical types described and defined here is a matter for separate discussion which, as I hope, is to be conducted later, as noted at the beginning of the article.

[13] The fact that this transcendental system is subjective, in contrast to the Kantian conception, will be discussed in the next section.

[14] This, of course, is in addition to rejecting systems that do not appear to them binding at all, even if they do not contradict their basic system.

[15] For challenges to the assumption, in transcendental arguments, of the unity of such systems, see, for example: 'The Impossibility of Transcendental Deductions', Monist, LI (1967), No. 3, 14., and also

'Transcendental Tendencies in Recent Philosophy', Journal of Philosophy, LXIII (1966), No. 19..

[16] In his book 'Faith, History and Values,' Akademon Press, 1982, in the final essay.

[17] As for the change of a basic system of thought, see the discussion by S. Körner, 'Categorial Change and Philosophical Argument,' Proceedings of the Israel Academy of Sciences and Humanities, vol. 4, no. 8, 1970. The above discussion is by no means exhaustive, and in my view suffers from several incautious arguments. His claim there that categorial change is not possible 'from within' but is possible 'from without' requires a prior definition of the relative status of the two categorial systems, and of whether, when there is overlap between such systems, they can still each be defined as a system in its own right and yet be called different. Such a situation contradicts his own definition of a categorial system, but this is not the place to elaborate. In my opinion, his error lies in the very conception underlying his argument, namely that a change of a categorial system also takes place within the framework of such a system, subject to the formal definition he there gives (although he too agrees there that it is not likely that an explicit description of this system can be given that would make it possible to foresee categorial change before it occurs). As I argue here, such a change cannot be given a formal description within a system of this type, neither 'internal' nor 'external.' This is what I called in the body of my remarks "the second-order transcendental argument".