A Defense of Ethical Stability
When discussing ethics, we have an inherent predisposition towards favoring the absolute primacy of reason, in specific the rationality of our current views of what a preferable action consists of. This position seems like an obvious one to take – if we have a particular view of what the right and wrong thing to do is, then it would behoove us to do the good as opposed to the bad action (with “the good” being defined as per my previous article “The Question of Liberalism”), so as to be a good as opposed to a bad person. Furthermore, if we are to take the desire to be and do the good as opposed to evil as the basis of our ethics, then we ought to also reject any action or desire other than the good. However, in this article, we will find that such a rationalistic attitude is in fact contrary to what we should actually pursue, instead arguing that certain viewpoints and attitudes must be insulated from arguments of whether they are good or bad. We will therefore conclude that one must coexist with certain negative actions or attitudes, so as to avoid greater evil.
In this article, we will first analyze the basis for the concept that we ought to desire greater goodness, looking at two particular bases for the good. Then, we will look at how these bases are dependent upon particular assumptions which, if they were to change, would lead to undesirable outcomes. From there, we will look at how different actions that seem to be implied by a desire for a better conception of the good can actually lead to those undesirable outcomes. We will then look at what in particular can be done and what safeguards can be employed to avoid such undesirable outcomes from occurring. We will then look at both of the possible results, the outcome created from solely pursuing good actions as well as the outcome created by implementing required safeguards, and look at why we ought to prefer the latter, more minor evil to the former. We will then close by examining the implications of our theory and a recapitulation of our argument.
Before we begin to look at how a constant application of our conception of goodness can lead to undesirable outcomes, we should look at what outcomes we desire from good actions. We will assume our basis is either that we wish to look upon our previous actions as good or that we simply wish to do the good as we currently define it for ourselves. In the first case, we wish to view ourselves as good rather than bad people, and our actions reflect that desire. Our desire for goodness, therefore, is sated if we are able to define our own actions as good as according to whichever definition of good we currently believe in. This perspective means that we do not only care about what perspective we currently have about our actions but also what perspective we will have in the future. So, if we can anticipate that we will regret or condemn an action in the future, even if we do not regret or condemn it right now, then we ought to not take that act, as our future perspective on our current action will be just as important as our current perspective it is now. We will describe this perspective as the Regret Basis for Goodness (or simply the Regret Basis for short), as it argues that we want to act good to avoid any future regrets about our current actions.
On the other hand, we also have a second basis for good action, one which takes goodness to be good on its own, separate from our particular perspective. According to this second basis, whether we do or will see an action as good or bad does not matter in determining whether we should take that action. Instead, our basis for acting good as opposed to bad is for the sake of being good itself, not for our own future pleasure at seeing that we had previously done a good thing. According to this perspective, good actions are simply superior to bad actions, as the term “good” seems to imply, and therefore, we should plainly act in whatever way we see as good. This perspective therefore takes the concept of goodness as a given, an axiom which does not change with our perspective of what it means. This perspective must assume that we currently know what the concept of goodness is, or at least, that we should act as if we do. If it did not assume we already know what the good is (or at least act as if we do), then we would not know which actions are good or bad and therefore cannot choose between actions, creating paralysis. Indeed, we couldn’t even contemplate any kind of actions necessary to leave that paralysis (such as thought about what goodness consists of) for fear that those actions would be bad. Therefore, to avoid a completely impracticable theory, we must assume that this approach focuses exclusively on endorsing whatever our current perspective of the good is. We will call this the Preference Basis of Goodness (or simply the Preference Basis for short), as it argues that the reason for our desire for goodness is simply our preference for what we have identified as goodness as opposed to badness.
Both bases of goodness are inherently hostile to shifts in perspective over what goodness consists of. Let us start by looking at the Regret Basis. For someone who believes in the Regret Basis, the worst thing possible that could come from their concept of good or evil is that they could view themselves and their own actions as wrong or bad. That does not simply include actions that they will take in the future but also actions they have taken in the past. Therefore, if their concept of a particular action changes from believing it to have been right to believing it to have been wrong, then having taken that action would be bad. Indeed, it would be just as bad as if they had taken an action that was intentionally bad, as in both cases they took an action that they now equally regret. This argument therefore implies that if a group of actions is condemned by a particular change in perspective (say, on goodness), then that is equally bad to similar actions being taken specifically to do what is now condemned. Therefore, if we come to later condemn a group of previously acceptable actions, that change in itself would be considered gravely undesirable by the Regret Basis. Therefore, a shift in perspective on goodness would be considered gravely undesirable as well, as it would lead to previous actions (taken under a different concept of goodness) demanding condemnation – and, more importantly, being regretted.
The same argument is true from the perspective of the Preference Basis, although for different reasons. From the perspective of the Perspective Basis, any view of the good other than the one currently held by a given person would be considered a source of evil, as it would cause actions other than those which one’s current perspective deems good. As a result, the concept of taking a future action which would be considered bad according to one’s current view of the good, even if one will eventually endorse a different concept of the good, would be considered by the person currently as something negative, and therefore as something they must stop from occurring. Thus, the concept of a shift in perspective on the idea of the good, which would imply many different actions that one’s current conception of the good would not want, would be greatly undesirable, as it would lead one to take actions different from those currently considered desirable. Therefore, in the current point of time, one would want to stop any actions contrary to their conception of the good from being taken, even if they would not object to such actions at the future point in time when they would actually act. So, we should reject any actions which could cause one to act in a different way from what we currently believe is good. A shift in our perspective on what the good consists of would be such a shift and therefore wrong, according to the Preference Basis.
Thus, we have found that both bases for the good demand one’s concept of the good remain stable and not significantly change. However, this argument goes quite a bit further than simply opposing a shift in perspective. In fact, using either basis of good, it is undesirable to simply deny one’s knowledge of what goodness entails. From the perspective of the Regret Basis, dogmatically insisting on one’s idea of goodness even after finding it to be wrong would lead to a great deal of uncertainty, as their applied view of goodness would be different from what they in fact believe would be the most reasonable concept of goodness. Thus, instead of simply questioning our actions later and regretting them, we would be forced to regret our actions we take when deciding what is good and bad. Therefore, we would regret the basis of all our actions – and therefore, whether or not we regret (or would regret) every action which we take, we would be uncertain at all times if we ought to regret every action we take.
The position of simply denying what we rationally believe to be the correct definition of goodness is even more impossible from the perspective of the Preference Basis. To the Preference Basis, the correct meaning of goodness is the fundamental basis for acting a certain way. So, while we could technically deny a better meaning of goodness than the one we currently use, then we would have no reason to act in the way we currently do with our current concept of goodness, as our reason in the first place for acting as we did was to make our actions correctly reflect the concept of goodness. Therefore, we would be self-aware to know that they were not acting in what way we considered good, and therefore, with every action we take, we would be acting contrary to how we believe they ought to act, contrary to the fundamental idea of the Preference Basis, that we inherently have good reason to act how we ought.
From both perspectives, not only must the actual affirmation of any assertion contrary to our current view of goodness be resisted, so must a great deal of actions which come prior to one’s rejection of our current viewpoint. As such, if a rejection to our current moral viewpoint never comes up, or if it remains weak enough to where we believe our current viewpoint is superior to the objection, then there is no basis for any uncertainty or disagreement with our current moral viewpoint, and therefore, we face none of the same problems as we would if we simply rejected an assertion about the good that we otherwise might find reasonable. However, this calls for a great deal of shifts in our actions so as to avoid anything which could cause us to reject our current view of the good. This not only requires a shift in our thoughts but also our actions, as putting ourselves in different situations could lead, even if we were to make sure to avoid certain thoughts, to other viewpoints which would have just the same effect, whether we wished them to or not. Clearly, fundamentally different actions (even different actions from what one would normally consider good) are required so as to avoid the rejection of our prior idea of the good.
At this point, we must ask some questions about how exactly we can come up with what actions we ought to take or not take to avoid the rejection of our previous idea of the good. We will consider several possibilities. First, we will consider a safeguard built out of previous rejections of our previous moral view, which demands that we avoid whatever actions have previously led us to reject our viewpoint. Next, we will consider rationally examining our current view of the good and analyzing all the weakest parts of it, demanding that we close off any discussion of the most vulnerable parts of our idea of the good. Finally, we will look at and endorse the idea that the basis of our actions ought to be taken from perspectives which have previously managed to be stable, rather than by an empirical look at mistakes we have made or rational thought.
The first safeguard one might have is to adopt an empirical analysis: that we can simply change our view of what actions should be taken based on previous rejections of our concept of the good. According to this perspective, we frequently change our view of the good. Therefore, we can, over time, find the actions that lead to those changes. By doing so, we can then prohibit certain types of actions, therefore eventually leading to a prohibition of most kinds of actions which would lead to the rejection of our current morality. In the future, when we might otherwise act in a certain way that could lead to our rejection of our current concept of the good, we would then recognize the situation and prohibit whatever action we might otherwise take, thereby protecting our previous concept of the good. While we could never prohibit every single action which would lead to our concept of the good being replaced, we could prevent enough to find ourselves in a much more stable perspective than would otherwise ensue.
In spite of seeming reasonable, however, this empirical perspective has several main problems. The first problem is that it is very rough. By its own admission, this perspective cannot properly safeguard anyone from most things which would lead to shifts in morality. As such, any other approach which could produce any broader an insurance against shifts in morality would be preferable. On top of that, this perspective would lead us to reject certain moral actions in an unconscionable manner. We would find ourselves justifying bad actions undertaken for the purpose of avoiding moral shifts – not based on our moral obligations or even our duties, nor simply by habit – but by a conscious statement of, “What I am doing is wrong, but I will do it to avoid changing my moral position.” Such a statement is quite obviously objectionable and would lead to one’s moral condemnation of themselves quite often (as all actions which one would notice repeatedly leading to shifts in one’s worldview and would therefore be prohibited would be actions which one repeats, at least some would likely be actions often taken). Finally, and most importantly, this perspective does not properly consider what actually leads to shifts in moral opinion. If an action leads to a shift in opinion, that action likely has something to do with the resulting shift in opinion; for instance, the shift in opinion would be a condemnation of that action based on experiencing its negative consequences. It is likely, therefore, that an actor’s change in action would lead to their moral opinion shifting such that they need not shift it anymore, and therefore, that the actor needn’t make any more changes to their moral position to avoid further moral shifts. For instance, if someone learns that murdering is wrong by murdering and therefore avoids it in the future, they need not avoid positions in which murder might be a possible option, as they have already realized that murder is wrong. To prohibit actions in which the idea of murder could occur to the actor at this point would simply be spurious, as they would not commit it even if they were put in that position. On the other hand, very few actions which would lead to future shifts in opinion would likely be prohibited prior to the occurrence by this strategy, as the only actions which would be prohibited are those which have already been found to be bad. If someone had not previously murdered, for instance, they would have no way of knowing, by this perspective, that murder would cause their morality to shift. As such, the most important part of what this strategy is trying to accomplish, avoiding future shifts in moral opinion, would not be accomplished. For any action that does not exactly resemble a previous action, one would have no way of figuring out if taking that action would lead to a shift in moral opinion. Therefore, this perspective would be completely ineffective in regulating many actions (those which one has not previously enacted) and overly effective in others (those which one has previously enacted).
The second perspective we would consider adopting is a rationalistic one which bases our actions on sealing off thought about any of the most unreasonable parts of our moral position. According to this viewpoint, we ought to, in adopting a viewpoint, look over all its best and worst parts. After recognizing which parts of that viewpoint were weakest and most clearly problematic, rather than remedying them, we ought to simply avoid thinking about them. In the future, if we recognize at any point that we are taking a position that could lead to a moral shift, we ought to reject anything we are currently thinking that may be leading to that conclusion, and instead think or act in a different manner. As a result, we will no longer have to wonder if our positions are right and wrong, and the prospect of our moral opinions changing will no longer be a problem.
This possible approach has two main problems. The first is that, in trying to defend a moral position, it would lead to its own demise. In order to effectively prevent addressing the main problems with our positions, we first would have to identify those main problems. But if some of those problems are compelling enough that the only way to overcome them is to avoid discussing them, then they would be problems that couldn’t be simply surmounted – they would represent fundamental problems with one’s ideology. Those problems wouldn’t simply be rendered impotent because one’s intention in identifying those problems changed to trying to root out all possible problems. But upon identifying those problems, we would therefore already be required to reject our viewpoint. If we refused to reject it at this point, then the problems would be just the same as in our previous discussion of what occurs if we dogmatically reject any logically founded alternative position to our current position. The second problem with this approach is that it only takes into account one’s current thoughts and beliefs about their particular moral system, therefore not taking into account problems with one’s ethical system which are not currently understood by the actor. While some actions which cause one to reject their moral position can be predicted by the actor before the actor actually takes them, others are not as obvious, particularly actions which do not directly cause one to change their moral position but rather those which force one into a situation in which further actions absolutely must be taken which then cause a shift in moral opinion. For instance, it might not be predictable to an addict that their first usage of tobacco will cause them to then become addicted, and addiction might lead them to perform immoral actions to secure tobacco, yet one ought to be able to predict how that immorality might come about ahead of time. Predicting future actions becomes yet more difficult when looking at actions that one has never taken, which likely means that this rationalistic strategy would fail to safeguard against many possible future actions. Finally, as this rationalistic strategy does not prohibit actions but rather prohibits thoughts, it allows the actor to think about exactly the negative problems with their beliefs that they are trying to avoid. Further, as previously established, the rationalistic strategy demands the actor think about and recognize each and every problem with their moral theory. And, when one thinks a thought that will lead to them recognizing a given problem with their moral theory, they already know how that thought will lead to the problem with their moral theory and what that problem is. So, if one uses the rationalistic strategy, they are required to simply, upon thinking a particular thought, remind themselves of a significant problem in their moral theory and tell themselves that they should not consider it. As a result, this approach creates the exact problem we diagnosed earlier when looking at an unnuanced rejection of any contrary positions to our own: that we will be required to know that our beliefs are wrong while still acting as if we believe them.
Our third possible approach is significantly superior to either of our first two. According to this attitude, certain groups of positions and actions are more stable than others. What that means is that, for some reason or another, there are some basic concepts of the good which do not often face significant rebellion or change from those who believe in them. This fact is established empirically – there are certain viewpoints which have existed for a long time with very little apostasy. Those viewpoints, therefore, are those which can be trusted to be internally stable and to avoid significant shifts in perspective and to have their own built-in safeguards against such shifts. Therefore, one need not consciously try to oppose certain viewpoints but rather simply confine oneself to pre-made viewpoints which happen to avoid moral shifts.
This perspective has many positive traits as opposed to the previous ones. Perhaps most importantly, taking this position does not require us to consciously act contrary to our moral views. In fact, the moral views which this viewpoint espouse will often, if not always, include safeguards against moral shifts as positives and will condemn those actions which will lead to moral shifts. So, when one consciously acts in a way that avoids ethical shifts, one does so because they believe doing so to be part of the good. This is significantly more acceptable as a position than if we were to demand people take morally wrong actions for the sake of moral stability. It also avoids difficult questions of moral tradeoffs of different actions. For instance, if one were to weigh two actions, one which was morally evil but would avoid moral shifts and one which was morally good but might lead to a moral shift, there would be no clear way to decide between them. By avoiding taking actions solely for the sake of increasing one’s moral stability, we can avoid such difficult questions. Another positive is the fact that this viewpoint does not confine its implications to a particular repeated situation or thought but rather extends to many possible situations and thoughts. Because it is based upon moral guidelines which one can apply to different situations, this viewpoint makes sure to not either over or under regulate action. Rather than only prohibiting universally negative actions within certain contexts, if certain kinds of actions generally cause moral shifts, then those actions will generally be prohibited. And, rather than universally prohibiting actions which might allow stability in some contexts and not in others, this approach will only prohibit those actions within the particular context in which they cause instability. (Of course, this will not be a generally perfect system, and there will be problems in many places, places where certain things are banned which should not be and places where certain things are not banned which should be. However, in general, it will likely track well to what should and should not be banned.) This perspective can be contrasted with the alternative approaches, which are both extremely arbitrary in their bans, either creating only universal bans on actions (and therefore banning many actions in contexts where they do not pose any threat to one’s moral position) or creating only situational bans on actions (which do not properly ban things which are universally destabilizing).
However, if our position is not already a stable system of thought, such a position does require a particular rejection of our current moral views. In fact, it requires a perhaps more capricious rejection than any other possible rejection, one which actually requires us to take a position contrary to our morals. It is therefore quite questionable, without any clear process to avoid any major ethical problems, to take the position that we should endorse this particular view rather than simply allowing for changes in our ethics in the future. How can one say, with any amount of certainty, that holding a position one thinks is wrong for the sake of ethical stability is somehow superior to consistently changing one’s position regarding what one thinks is right? In fact, it seems that this position arrives at exactly the problem that we diagnosed earlier about taking a position one knows to be wrong – we are just as uncertain and our actions just as factually bad as if we held our original position.
While this seems like a compelling argument, and indeed, it does require some changes in what we have posited thus far, it does not require the complete rejection of our hypothesis. Indeed, for someone who does not currently hold a position which follows the rules we have put forth for a stable concept of the good, mere stability cannot be enough to demand their support for a particular concept of the good. However, this does not apply to everyone. First of all, some people already have a stable concept of the good. For them, this approach merely demands that they do not undermine their current concept and exchange it for another. Secondly, one can change their concept of the good forcefully if they so desire. As previously described in “Apostasy and Paradigmatic Shifts,” one’s paradigmatic views of the world can be changed by force if one acts in a particular manner over a long enough period of time. (In that article, we argued that if a particular paradigmatic view was self-proving, then merely taking actions and thinking thoughts based upon that view can eventually lead to one adopting that view.) Therefore, changes in an epistemic position of what constitutes goodness can be changed simply by acting upon certain principles consistently. As such, our view of goodness can be manufactured without requiring any cynical actions based on a desire for moral stability at the expense of moral goodness. However, what this does demand is that our ethically stable views not simply be based on a particularly stable concept of the good but also on an epistemic view of what goodness consists of (as well as an epistemic view on the world, so as to retain consistency), to avoid any kind of capriciousness in our shifts in views. And this then demands (as we will discuss later) apostasy.
So, then, we can discuss exactly what the implications of our view are for ethics. The first has already been foreshadowed: it demands that we keep to stable viewpoints. Now, this is not the same as demanding that we follow a particular viewpoint or that we follow a singular viewpoint no matter what but merely that any changes in viewpoint be guided by a view of how we can retain stability in our actions. Therefore, changes in viewpoint can only be allowed to occur between stable viewpoints. Secondly, such viewpoints should be tied somehow to our epistemic and paradigmatic viewpoints, so as to avoid the self-defeating concept that we are only undertaking a change in our viewpoints for the sake of stability – we must have a reason for preferring the new viewpoint other than a cynical desire to avoid further shifts in viewpoint. This then requires that our viewpoints shift in concert with our views of what a good viewpoint consists of. This, of course, requires apostasy, as we have defined it in “Apostasy and Paradigmatic Shifts.” That is to say, one’s viewpoints must change based not on a logical shift in one’s views of the world but on an extra-logical shift in what one views the basis of logical views to be.
The second implication of our argument is that the stability of one’s views ought to be valued quite highly when we consider different moral views, perhaps more highly even than possible internal contradictions. If a particular viewpoint requires some kind of contradictory statements for the sake of ethical stability, such a viewpoint can, according to our argument, be superior to a more internally consistent viewpoint that will eventually lead to its own demise. Indeed, if a particular action or basis for action is defined as negative by another thought (and is therefore contradictory with the other thought), that is undesirable and causes self-condemnation, which we have previously determined to be quite undesirable. However, if all previous actions are going to be perceived as negative by a future viewpoint, then that is a much greater threat than the mere condemnation of a single action which one has previously taken, as is required by a contradiction. Consequently, a contradiction (particularly a minor one) is a comparably unimportant detail and is preferable to the alternative problem of a general condemnation of previous actions. In addition, according to our doctrine of stability, simply overturning a contradiction would be a negative, as it would lead to possible undesirable consequences in the future, namely the condemnation of all one’s previous views.
And, of course, this approach has a great deal of pragmatic consequences, for it demands that we take entirely different positions from the ones which we would normally take. It demands that we act in a manner closer to views which are agreed upon over time – so as to retain stability – rather than views which are comparably true or seemingly true. It demands that we change our positions only to other positions which are equally or more stable, and that we do not simply change our moral positions to iron out any seeming contradictions or problems in our current views. This argument therefore demands that we take a much less absolute approach to our ethics. Rather than demanding a particular ethical system be perfect, we ought to simply compare a group of stable ethical systems and then, after choosing one, only change our viewpoint in the future if we come to believe that another stable viewpoint is significantly superior. And, perhaps most importantly, it demands that we value stability above nearly all other factors of an ethical system and that we allow for internal disagreements within our viewpoint in order to avoid significant shifts in our ethics that would ultimately be contrary to what we desire.
Finally, we will end this article with a recapitulation of our argument. First, we defined two different ways to ground our desire for goodness, one which says that we do good to avoid regretting bad actions and the other which says we do good because good is, by definition, better than bad. Then, we looked at how both perspectives imply a preference for one’s views on goodness not to change. We then saw that what would seem like the implication of our previous point, that we should simply deny ethical change contrary to our current beliefs, even if we recognize that those ethical changes would make us better people, doesn’t in fact hold up, instead finding that we must change our actions in order to prevent changes in our views on the good. We then looked at several ideas of how we could change our actions in this way, finding all to be insufficient except for an approach in which we adopt an ethical framework that people have previously found stable and hold that one to avoid shifts in our views. We then considered an objection to that approach and found that one’s ethical views must be tied to their epistemic views, and that, therefore, to change one’s ethical views to be more stable must also include a shift in one’s epistemic views to make them more compatible with our new, more stable ethics. From that conclusion, we found that the disagreement between different views within a particular moral perspective must be tolerated as a requirement for stability so as to avoid the creation of fundamentally unstable views, which would lead to disagreement with all one’s current moral views rather than merely some.
Of course, this discussion of morals, which argues that tolerating some moral disagreement and the moral inferiority of some actions to others must be accepted to avoid shifts which cause one’s morals to lose all power over future actions, is closely allegorical to the dilemma which faces the subject of our next article: the state.
