The forces which restore, or endeavor to restore, disturbed equilibriums in nature, achieve, or endeavor to achieve this, with counteractions (such as a higher reproduction or reproductive capacity in the event of a diminishing population). But each counteraction has to be as unneutral as the disturbance it is meant to offset in order to be wholly effective. The striving for maximum entropy in systems which can take their own course has neutrality as a result, that is, a homogeneous concentration throughout a closed system, but the changes in concentration in the different parts of such a system are unneutral themselves. The process in which the change in an external condition of a chemical system is leveled as much as possible is a reaction process which is basically unneutral itself, yet aimed at neutrality, that is, equilibrium. Induction currents work against the change by which they came into existence, yet these forms of electricity are as unneutral as the change which caused them. The principles of conservation of momentum, of rotation, of charge and of mass and energy imply that increase-catenary neutrality, or the state in which the total amount of these quantities does not alter, can only be maintained with an equally unneutral counterchange for every alteration. These are all examples of a neutrality needing a certain kind of polarity.

Striving for neutrality in the event of polarity therefore requires increase-catenary polarity, that is, polarity with respect to the increase-catena of the original catena whose neutrality is strived for. If good is equated with of superior normative value, the neutrality strived for is 'good', whereas all unneutralities concerned are 'bad' on the principle of catenated neutrality. But if the neutrality strived for is 'good', and if right is equated with promoting the good, the act of striving for neutrality is 'right' even tho such an act requires comparative unneutrality, that is, a form of 'badness'. On the other hand, an act of striving for polarity may require the same degree of polarity, and thus be equally bad, but wrong as well. Strictly speaking, the principle of neutrality solely determines what is 'good' or 'bad' on the performatory level, whether in the sense of what is superior, or in the sense of what actually or probably promotes what is superior. Without further refinement it cannot even determine what is 'right' in that it promotes neutrality or intends to promote it, for what promotes neutrality is unneutral in itself. On the decision-theoretical level, however, 'right' is indeed what is decided to promote a neutral cause, and 'wrong' what is decided to promote an unneutral cause. (Note that in the context of this chapter we must always read "prima facie right" and "wrong", or add insofar as the norm of neutrality is concerned.)

Even when the distinction between goodness and rightness is made with regard to the neutralist evaluation of the ground-world, the principle of neutrality needs at least to be refined or interpreted in such a way that the polarity which is inherent in the striving for (the greatest chance of) neutrality cannot itself be the object of a right action. Otherwise it would even be right to offset and upset acts or processes aimed at neutrality. Given that this is not right, our interpretation of the principle of neutrality is of the neutral, unless type: one should not aim at polarity, unless such polarity serves a neutral purpose. This is tantamount to saying that neutrality is normatively superior, but that some aspects (first of all the auxiliary ones) have to be treated differently on the basis of this principle, because such a distinction between aspects is relevant to a neutral end.

What nonauxiliary (or 'notional') aspects can or must be treated differently if the neutrality of a certain catena is indeed a legitimate goal? To answer this question we should, among other things, look at each catena's position in a derivation system. Let us call the catena with the neutrality which can be chosen as an end in itself "catena A". Obviously, if the neutrality of A is a legitimate penultimate end (with neutrality per se as ultimate end), the neutrality of the increase catena of A can only be a legitimate end as well if the state of affairs in question is neutral from the angle of A. The superiority of increase-catenary neutrality is therefore a conditional one, given that an original catena takes precedence over a derivative one. What holds for the neutrality of the increase- or differentiation-catena, holds for the time-differential catena of A: rest or constancy are normatively superior, unless movement or change over time promote neutralness in terms of the original catena. This is the relevant distinction we draw between two types of constancy and two types of change over time -- 'relevant' that is, in respect of an original catena.

Now, what is said to 'promote' neutralnesss may still be a question of physical modal conditions, or a question of catenical relations. And as we are in this context concerned with catenical relations, we can be more precise. We must, then, distinguish between neutrality-differentiation and -differential catenas with an aspectual value 0 for the differentials, and positivity- or other differentiation and differential catenas with an extremely positive or other unneutral aspectual value. The special relationship is, then, between the neutrality of the original catena and the positivity of both the active and the passive neutrality-differentiation catenas and of the neutrality-differential catenas. If we do not refine the principle of neutrality and accept that positive, active and passive, neutrality-differentiation and neutrality-differential change are 'bad', then it will not follow that we should abstain from this kind of badness. We could also refine the principle's formulation and say that it does not apply to the positivities of the neutrality-differentiation and -differential catenas of a catena whose neutrality is unconditionally superior. Whether we opt for the one or the other approach, it does not affect the content of the norm of neutrality. What does affect this content is the catenical distinction which has to made between different aspects.

Sometimes positive polarity can be a sign of neutralization, and negative polarity a sign that something is becoming, or is made, (more) unneutral. Just as it does not make sense that neutralization would be inferior to increase-catenary neutrality, or that neutralization would be as inferior as polarization, so it does not make sense to believe that something that is merely a sign of neutralization would be normatively inferior. The neutrality of such a catena would therefore not be superior. A possible example of a catena which has such a signal function is the happiness catena. This catena will be discussed when we examine the concept of the well-being of sentient beings and the principle of beneficence.

The principle of neutrality does not lay down what empirical value of a quantity which can assume three or more values is the neutral one and corresponds to the catena value 0. However, given the neutrality of the original catena A, and granted that this neutrality is of superior value, another neutrality cannot be a penultimate end if it belongs to a catena which has been factitiously derived from A. (See I.2.5.3 for the difference between factitious and nonfactitious derivations.) This includes all modulus-catenas. Hence, it cannot be argued that, for example, moderateness is the opposite of extremeness, and that therefore it would be the neutral limit-element between moderateness and extremeness which were normatively superior. The principle of neutrality can solely hold for nonfactitious derivations, basic catenas included. A special type of nonfactitious derivation is, then, the bicatenal bivariant comparative catena for which the neutral value corresponds to equality. Since this catena is neither factitious nor a differentiation or time-differential catena, the principle of neutrality is unconditionally applicable to it. Bicatenal bivariant equality is normatively superior to the concatenate inequality, regardless of the original catena values concerned. This derivative principle is the principle of equality; of catenated bicatenal bivariant equality, to be precise. We will review this principle and traditional objections to it in a separate division of this chapter (namely 3.5).

There is at least one more connected series of predicates with regard to which the principle of neutrality does not seem to make sense from a purely conceptual point of view. (Other points of view would require other normative, ground-world principles and would mar the fundamental character of the present analysis.) It is the temporal auxiliary series. If someone were to say that existing-now is normatively superior to existing-in-the-past and to existing-in-the-future, there would be no way to promote this form of the good. The course of time itself cannot be influenced, and things which exist in the future, will exist at a time which is then the present moment. (Or --as a sage of old had it--, 'there is no limit to the present moment'.) Time is an aspect different from all others, and people are even willing to accept the asymmetry of time in that they believe that they can only go forward in it, and not backward. Such asymmetry causes irritation and is being challenged by theorists in search of a ubiquitous and eternal, fundamental symmetry in nature. On our neutralist model it would be nice if the belief in the asymmetry of time proved to be mistaken (provided, of course, that we have not to put up with unneutrality elsewhere). But if there is any factor that 'behaves' differently, it is the factor time, and therefore it is correct that the principle of neutrality does not apply to such an auxiliary predicate. It is also time-differential catenas it does not unconditionally apply to, not any of the other differential catenas.

It should now have become clear why the principle of neutrality has to be refined or interpreted in a relevantist way. There are distinctions to be drawn between certain kinds of aspect which are either relevant from the perspective of the superiority of neutrality itself or from the perspective of the adoption of the principle of neutrality. To sum up: the principle of catenated neutrality, whether its formulation is refined or whether it is interpreted in a relevantist way, only applies to:

1. connected series of nonauxiliary, nonevaluative, primary predicates --therefore it does not apply to normative, modal and temporal, auxiliary series with auxiliary predicates like goodness, probability and being-in-the-future--
2. catenas which are not factitious --therefore it does not apply to modulus-catenas and certain other difference catenas--
3. catenas with respect to which a primary thing does not need to have an unneutral predicate, if it is to strive for, attain or maintain the neutrality of an original catena --therefore the neutrality of an original catena takes precedence over polarities of differentiation and time-differential catenas--
4. catenas of which one of the monopolarities is not a sign of neutralization with respect to another catena --therefore it does not seem to apply to the happiness catena as we will see later--.

It is thinking of the last two types of catena that one can say "neutrality is good, neutral-directedness is right".

THE   CATENICAL   DAO
 
By our stillness we shall be supreme.
By our movement we shall bring it about.