Consequentialism tells us that we should aim for the best result. But what is the best result? A common way of comparing one possible situation to another is by assuming that each individual life has some value associated with it. That value may be happiness, or quality of life, or whatever the ethical code we are considering considers to be most important. We then sum this value over all the people living, and take this sum to roughly measure how good the result is in comparison to other results (in other words, a higher sum is better than a lower one). Utilitarianism is one example of a philosophical position that measures results by this kind of standard, as it recommends the outcome that would maximize total happiness.
Unfortunately, this way of comparing possible outcomes has a serious defect. Consider a society, A, which has 20 people whose lives each have 100 units of value (happiness or whatever). A second society B, which has 40 people whose lives are each valued at 51, is better than society A by this method (since the total is greater). And society C, with 80 people whose lives are each valued at 26, is better yet. And so it is possible that the best possible outcome is one in which there is a vast number of people, each of whom have lives that are barely worth living. Clearly this is ridiculous, and so evaluating outcomes by summing the value of individual lives must be flawed in some way.
One possible alternative is simply to consider the average value (instead of the sum) as the relevant measure. Such a method doesn’t lead to the absurd conclusion outlined above; it suffers from a different problem. If we subscribed to the average value method we would conclude that a society of one happy individual is better than one consisting of millions of people almost as happy, and conversely a society composed of one individual suffering is worse than one composed of a million people suffering almost as much. Such results are just as unacceptable as those advocated by the total value method, and so it can’t be the correct way to determine which outcome is best either.
Between these two extremes there are numerous other partial solutions, for example we might propose that there is some minimum value a life must have before it counts towards making a solution good, or that there is some maximum quantity of lives that can contribute towards the goodness of a solution. Besides being unmotivated by theory, these solutions lead to their own varieties of unacceptable outcomes, which I won’t detail here since Derek Parfit has already done an exceptionally thorough job in his book Reasons and Persons.
What all these proposed methods for evaluating the goodness of an outcome have in common is that they judge the outcomes based on how well individuals are doing. This I think is the fundamental problem. Earlier I have argued that ethics should be viewed as a practice whose primary beneficiary is society (for example, here). This kind of view can help us determine which outcomes are best without falling into the problems encountered by the methods we considered above. What kinds of outcomes are good for a society? I would judge how well a society is doing based on two factors: how long it will survive and to what extent it is thriving. Societal survival should be a fairly self-explanatory criterion; certain societies (outcomes) will use their resources better than others and hence be able to sustain themselves longer. Of course by itself societal survival would lead us to conclude that the best society is an unchanging hunter-gatherer civilization, but fortunately the other criterion, how well the society is thriving, pulls us in the other direction. Societal thriving is a measure of how fast the quality of life is increasing for the members of that society and how fast technological advances are being made.
To compare various outcome we do not multiply or add some measure of how well they are thriving and the number of years they will survive; this would lead to the same kinds of paradoxes. Instead what we should do is consider what would happen if the societies created in each possible outcome were competing with each other. The one that would outlast the others is the best outcome. So while the hunter-gatherer society may have lasted longest on its own it would be absorbed by a society that was thriving to a higher degree, and thus it is not the best outcome despite its individual merits. On the other hand, a society that exhausts all of its resources in a short period of time won’t be the best, even if it is thriving in that time, simply because it won’t be around long enough to absorb the others. This kind of thinking also shows how we can determine how many people are best (as the other methods seemed to recommend as many as possible or as few as possible in some cases). Here the best number of people is one that allows the society to thrive, but doesn’t deplete its resources. Clearly the exact number will depend on how many resources are available, the level of technology, and how much space is available, but it should be obvious the best number of people will neither be as many as possible nor as few as possible.
Some may view this method of judging various outcomes by their survival when pitted against each other as somewhat harsh. You might think that this would lead to us calling the most aggressive and vicious societies best. This is not so, not because we invoke some special rule to bar them, but simply because a society that is ethical, at least internally, will simply do better in the long run than one is not. This is an idea I have discussed earlier. Another possible problem with this method is that it seems to apply only to situations involving choices that have wide reaching effects. After all how can we compare a society that involves one person picking up their trash at a specific time to another that is exactly alike except that the trash is left unpicked up? The solution is to “blow up” our reasoning in order to make the difference noticeable. Instead of pondering just my choice I should ask myself “what kind of society would result if everyone behaved / reasoned in this way, and would it be better or worse than one in which people didn’t?”, and now the possible outcomes will be noticeably different, allowing us to successfully compare them. Or we could consider how our decision impacts other people, and thus the wellbeing of society as a whole. Again, something that I have detailed earlier, and which leads to the same conclusions as the more “Kantian” method, although more precisely.