What Reasons Are There to Focus on the Long Term?

Original text Luciano Carlos Cunha · Volume IX, p. 32 of the PDF edition · English translation

4.1. What is longtermism?

The word longtermism³⁷ is used to express different ideas, depending on the author. Sometimes it is used to refer to a goal. At other times, it is used to refer to a practical focus for trying to achieve that goal³⁸. Below, two kinds of goals (one longtermist and one short-termist) and two practical focuses (one longtermist and one short-termist) are compared.

• Longtermist goal: aims to have the best possible impact on history from now until the end of time. This goal necessarily implies giving equal weight to every moment in time (that is, it implies temporal impartiality).

However, two people who agree on the longtermist goal may disagree about what we should focus on in practice to best achieve it. Thus, there are at least two kinds of practical focus for achieving the longtermist goal:

• Longtermist practical focus: to have the best possible impact on history from now until the end of time, we should focus on the impact our decisions will have on the long-term future.
• Short-termist practical focus: to have the best possible impact on history from now until the end of time, we should focus on the impact our decisions will have at the present moment or in the near future.

Now consider the short-termist goal:

• Short-termist goal: aims to have the best possible impact on the present or the near future. This goal will give less weight to (or even entirely neglect) the distant future (that is, it rejects temporal impartiality).

Since the short-termist goal does not seek to have an impact on the long-term future, there is only one kind of practical focus for achieving it:

• Short-termist practical focus: to have the best possible impact on the present or the near future, we should focus on the impact our decisions will have at the present moment or in the near future.

We thus have the following possibilities:

• Longtermist goal with a longtermist practical focus
• Longtermist goal with a short-termist practical focus
• Short-termist goal with a short-termist practical focus

4.2. What will we see in this chapter?

In section 4.3 we will look at the central reason for adopting a longtermist goal. In section 4.4 we will look at arguments for adopting a longtermist practical focus. In section 4.5 we will look at the main objections to the longtermist practical focus and how these objections can be answered.

4.3. Why adopt a longtermist goal?

The central reason for adopting a longtermist goal is this: if what matters to us is the good of all the beings we may affect, the question we ask will not be "how can we have the best impact on the present world or the near future?" but rather "how can we have the best impact on history from now on?". A more detailed defense of this idea can be found in Chapter 2.

4.4. Reasons for a longtermist practical focus

As we have seen, the practical focus on the long term holds that, to have the best impact on history from now on, we should focus on the long-term consequences of our decisions.

The main reason for the practical focus on the long term is that, unless there is a mass extinction of sentient life, it is very likely that there will be far more sentient beings in the distant future than in the present or the near future, both diachronically and synchronically:

• Diachronically because the future may be immensely long. The individuals alive now or in the near future are very few compared with those who will live in the coming centuries, millennia, eons, and so on.
• Synchronically because the population of sentient beings at each moment may increase. For example, space colonization could multiply them by many orders of magnitude³⁹.

In short, depending on how we decide now, we will positively or negatively affect a gigantic number of beings who will exist over the course of the future.

Let us consider an analogy to better understand the practical focus on the long term. Imagine that someone has to choose between avoiding a certain suffering now or avoiding a certain suffering twenty years from now. Imagine, however, that the suffering that would occur twenty years from now would be much more intense and prolonged. If the person chooses to avoid the suffering of the present moment, they will make their life contain more suffering than if they had chosen the other option.

Analogously, if what we aim for is the best history of the world from now on for all sentient beings, we need to give equal weight to every moment. However, since the long-term future will contain far more sentient beings than the present moment or the short-term future, we have strong reasons to prioritize how our decisions affect the long-term future — this is the basic idea of the practical focus on the long term.

Interestingly, even if less and less weight were given to the good of each sentient being depending on how far into the future it will live, the number of sentient beings who may exist in the long-term future is so vast that we would still have to conclude in favor of prioritizing them, if our goal is the best complete history of the world. Thus, even those who reject temporal impartiality but still give some consideration to future beings would have to prioritize them.

Of course, how long the future will contain sentient beings and what the population of sentient beings will be at each moment is highly uncertain. But future beings vastly outnumber present ones in expectation (that is, multiplying their number by their probability of coming to exist), and this is enough for us to have strong reasons to care about them. This problem of uncertainty will be discussed below in section 4.5.3.

4.5. The debate over the focus on the long term

Below, some of the most common objections to the practical focus on the long term will be discussed⁴⁰:

4.5.1. Is it better to leave future problems to future generations?

One objection to the practical focus on the long term holds that future generations will be better placed to solve future problems: they will know which s-risks⁴¹ (that is, risks of suffering on a gigantic scale) are most serious and which strategies are best for preventing them. For the same reason, the objection concludes, we have an advantage in terms of knowledge if we focus on the present.

One possible response to this objection is to note that it is not guaranteed that future generations will know how to solve such problems. For example, if we leave it to future generations to investigate these problems, it may be too late to prevent the worst scenarios⁴². In

short, this response holds that the sooner we begin working on these problems, the sooner solutions to them will emerge, and that this will make history better from then on.

Moreover, even if future generations are better placed to solve such problems, this does not mean they will want to do so (since this will depend on how much consideration they give to sentient beings). So we should try to bring about both the capacity and the will to prevent suffering in the future.

It is sometimes claimed that the future will be better anyway thanks to moral progress and that, therefore, we need not worry. However, this is not guaranteed, since it depends crucially on what we do now to spread consideration for all sentient beings and on what continues to be done over time.

4.5.2. Does focusing on the present have more influence?

Someone could accept that the good of future sentient beings counts equally but reject the practical focus on the long term because they believe we live in a crucial era for determining everything that will happen from now on. For example, certain events that will probably occur in the twenty-first century (such as the emergence of superintelligent, autonomous artificial intelligences) could have a disproportionate influence from then on. On this basis, it could be argued that, to have the best impact on the history of the world from now on, we should focus on the present (or, at least, on the near future).

One possible response here is to point out that we do not know how much influence each of these possible events will have on history from then on. For example, we do not know whether these will be the crucial events, whether others still unknown will be, or whether change will be gradual, with no crucial events. If we focus on particular events and they turn out not to have much influence on the future, we will have wasted our efforts⁴³. Thus, it seems better to focus on investigating how our decisions can influence the future positively regardless of what happens. This can be done, for example, by investigating risk factors for s-risks and by planning broad strategies for preventing them.

4.5.3. The objection that our degree of certainty about the present is greater

Another objection to the practical focus on the long term points out that the longer the time frame, the harder it is to predict the probability that an event will occur. For this reason, the objection holds, we should focus on the short term, since in that case our degree of certainty is greater.

However, the degree of certainty should not by itself determine how much we ought to focus on a scenario. Another crucial factor is the magnitude of the scenario (that is, how good or bad it is). Depending on how bad a scenario is, it may still be worth prioritizing its prevention, even if we are more uncertain about whether or not it will occur. One way of taking both magnitude and degree of uncertainty into account is to multiply one factor by the other, and the result is called expected value (or expected disvalue, if the result is negative⁴⁴).

Let us look at an example. Imagine that we have to choose between preventing a negative event that could occur in the near future and preventing a negative event that could occur in the distant future. Imagine that we are able to estimate how negative each event would probably be, and that the near-future event would be -100 and the distant-future event would be -1000. However, imagine that the probability of the near-future event occurring is 80% and that, owing to uncertainty about the future, the probability of the distant-future event occurring is only 10%.

The expected disvalue of the near-future event would be -100 x 0.8 = -80. The expected disvalue of the distant-future event, on the other hand, would be -1000 x 0.1 = -100. Thus, according to the expected disvalue calculation, we would still have reason to prioritize preventing the distant-future event, even though the probability of its occurring is lower.

In short, what this response points out is that the magnitude of the long-term future is so vast that, even if there is great uncertainty about the future, it would still be worth prioritizing trying to influence it.

At this point it could be objected that the expected disvalue calculation could lead us to prioritize preventing scenarios that would be disastrous if they occurred, but whose probability of occurring is extremely remote.

There are at least two possible responses to this objection. The first is to acknowledge that it points to a problem with the formula for calculating expected disvalue, but to argue that the formula can be adapted to correct this problem. For example, instead of the factors magnitude of the harm and probability of occurrence having the same weight in the multiplication, the probability factor could be given greater weight.

The second response points out that there are strategies that tend to influence the future positively without requiring us to estimate the probability of the future turning out this way or that. Longtermist strategies fall into two broad groups: targeted and broad. Targeted strategies focus on preventing specific catastrophic scenarios. Broad strategies, by contrast, are designed to influence the future positively regardless of which scenarios occur. Thus, the objection in question could make sense only with respect to targeted strategies, but would not affect broad strategies.

One example of a broad strategy is to identify risk factors for s-risks and try to prevent them. We will see more about this below.

4.5.4. The objection that the future is so uncertain that we have no way of deciding

A similar objection holds that the long-term future is so uncertain that we have not the slightest idea how our present decisions will affect it. For example, our influence will be diluted by the decisions of the many future agents⁴⁵. The longer the future, the more is at stake, but also the smaller our influence. Thus, the objection concludes, the fact that the future is long is not enough to support the conclusion that we should focus on the long term.

However, there are things we can do now that tend to influence the long-term future positively even if we are highly uncertain about the future. For example, we can identify risk factors⁴⁶ for s-risks and try to

do things that reduce these risk factors. Risk factors are not themselves s-risks, but they greatly increase the probability that s-risks will come about.

The concept of risk factors is already widely used in medicine. For example, smoking, eating poorly, and not exercising are not, in themselves, diseases, but they greatly increase the probability of various diseases, even though they do not necessarily entail them. To know that it is good to avoid these risk factors, we do not need to know what each individual's health trajectory will be. Similarly, we can try to identify risk factors for s-risks and try to identify the actions that would help avoid them.

Some of the risk factors for s-risks are:

• Disregard for nonhuman sentient beings.
• Disregard for possible future nonbiological sentient beings⁴⁷.
• Demand for products and services whose production harms sentient beings.
• The prevalence of environmentalist values instead of concern for sentient beings.
• Human expansion and increasing technological power.
• Technological progress being much faster than moral progress⁴⁸.

Each of these risk factors will be addressed in more detail in Chapter 8.

4.5.5. The objection that focusing on the long term would mean neglecting the present

Another common objection to the practical focus on the long term is to point out that it seems to imply neglecting present or near-future suffering, since this is vastly smaller in magnitude than possible long-term suffering.

One possible response here is to point out that, once again, there are actions we can take that tend to influence the whole of history from now on positively (for example, promoting the moral consideration of all sentient beings), and that, therefore, there is not necessarily a dilemma between addressing the long-term future and addressing the present.

At first glance, this may sound contradictory. Someone might ask, for example: "But aren't you defending the practical focus on the long term, then?". The answer is that some of the most promising strategies for the practical focus on the long term (such as advocating the moral consideration of all sentient beings) also help improve the short term.

4.6. Conclusion

We have seen that the practical focus on the long term is one possible answer to the question "how can we have the best impact on the history of the world from now on?". We have also addressed some objections to this focus and the responses that have been given to them.

There seem to be good reasons in favor of the practical focus on the long term. However, another important conclusion here is that even those who reject this focus (and choose to give priority, for example, to the present or the near future) still have strong reasons to accept the longtermist goal, even if they do not accept the longtermist practical focus.