Existential risk

A relevant concept in relation to existential dimensions of sustainability has been developed and made popular by Bostrom and colleagues. An existential risk is a risk where an adverse outcome would either annihilate Earth-originating intelligent life or permanently and drastically curtail its potential [24].

Existential risk is about risk (unactualized states of affairs). Bostrom advocates a two-dimensional view of existential risk. His terminology is not fully consistent between [24] and [19], but the dimensions are similar in content. In [24], see Fig 1, the two dimensions are labelled scope (endurable–terminal) and intensity (personal–local–global).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Adapted from Bostrom 2002 (Bostrom’s original examples have been kept).

https://doi.org/10.1371/journal.pstr.0000119.g001

In [19] the labels of the two dimensions are instead severity (imperceptible–endurable–crushing) and scope (personal–local–global–trans-generational–pan-generational). It is probably the interaction between the temporal aspects of the two dimensions that has led to these differences between versions. [24] operates with a seemingly atemporal intensity-dimension from personal to global. However, in [19] personal is expanded to global and then to inter-generational and thus entails a temporal aspect (that is instead present in the scope-dimension of [24]–entailed by the use of ‘terminal’).

If, for simplicity, we stick with persons and times as the two main variables, existential risk–according to Bostrom’s perspective–is about all persons and all future times. As Bostrom [24] says, “an existential risk is one where humankind as a whole is imperiled. Existential disasters have major adverse consequences for the course of human civilization for all time to come.” Hence, unlike the risks of species extinction or irreversible loss of biodiversity in ecosystems discussed by Naess (and recently in the first chapter of AR6 [25], and [26]) the existential risk is about humankind (or other intelligent kinds), and is in that sense necessarily about global consequences for intelligent kinds. This is not without its problems. In particular, if existential risks–thus conceived–are perceived as more important than risks run by individuals, there is an ethical dimension to this that deserves close attention.

In addition, existential risk highlights the risk that permanent damage obtains. What threatens to be permanently harmed when existential risk is not eliminated is something on the kind/species level–and only certain kinds/species matter.

Another important and interesting feature of existential risk is that it may apply to both humans (or other intelligent bearers) and their capacities. Bostrom refers to these as different classes of existential risk. The capacity Bostrom has primarily in mind is technological capacity. One might be surprised about this, but for him:

“The permanent destruction of humanity’s opportunity to attain technological maturity is a prima facie enormous loss, because the capabilities of a technologically mature civilisation could be used to produce outcomes that would plausibly be of great value, such as astronomical numbers of extremely long and fulfilling lives.” [19]

Clearly, for Bostrom technological maturity is a means, perhaps a necessary means, to maximize value satisfaction in the long run (including the elimination of future existential risks). Thus, other values than not going extinct can be included as sufficient for running an existential risk.

Returning to the connection between existential risk and existential dimensions of sustainability, it seems that the following implication might hold:

If X is being sustainable X is not leading to existential risk

This is because loss of the kind of life or its potential that existential risk is about (intelligent life) is proof of unsustainability, on all but the weakest versions of sustainability (e.g. [27]). And it seems clear that the implication holds if we consider Naess’ ecological sustainability. However, the reverse implication does not hold. This follows for two reasons. First, from the fact that running an existential risk has to involve possible loss (permanent damage) but unsustainability normally has not. Second, because existential risk is reserved for threats to “Earth-originating intelligent life” but unsustainability is not necessarily limited to such threats. Farming can be unsustainable, too, whether or not it has consequences on species level of those kinds that count, and whether or not the loss it entails can be reversed (for instance by other ways of farming).

The concept of existential sustainability could potentially be used to close the above gap, securing that the following equivalence holds:

X is being existentially sustainable if and only if X is not leading to existential risk

In order to illustrate this potential relationship, an illustration is borrowed from Mendes [20]:

“Why is the question of the sustainability of the Technosphere so anthropologically dramatic? Because […] the Technosphere constitutes a Technocene and a Technotope. It is no longer just a technological infrastructure in which we live our lives, but an artificial habitat, from which we cannot escape, to which we have to adapt, which determines our life modes and styles, the condition of the possibility of the continuity of our species, and our civilization. In other words, the question of the sustainability of the Technosphere seems more and more a global catastrophe risk and an existential risk.” [20]

According to the present suggestion, Mendes’ question of sustainability is so dramatic because it is a question not of any kind of sustainability but of existential risk. Thus, according to the alleged equivalence above, it is dramatic because it is not existentially sustainable.

However, this does not seem exactly right. There might be activities that are not existentially sustainable but do not lead to existential risk. It is only if we require that existential sustainability always involve kinds (i.e. humankind and similar kinds) that the equivalence holds. As we have already indicated there are two interesting differences to ponder. These provide an opportunity to refine our concept of existential sustainability, or more generally: existential dimensions of sustainability.

First, in the above existential risk discourse the choice of adaptive capacity that existential risk focuses on is placed on a societal/global level rather than on, say, an individual level (which wouldn’t make sense given that discourse, but might make perfect sense if existential sustainability was in focus). Technological maturity has a value on the aggregated level (due to its capacity to promote meaningful, valuable lives for individuals belonging to humankind for generations to come.) No one can argue against the importance of such a capacity. And that it has a place in sustainability thinking is beyond question.

It is less obvious that existential sustainability (a) has to be about adaptive capacities at the social/global level, and (b) only concern values that make our lives meaningful. Arguably, we could speak of existential sustainability also when individuals are concerned. Several uses of “existential sustainability” in theological/religious contexts concern the individual and her capacities rather than anything on an aggregate level. Moreover, these capacities need not only be of instrumental value. Finally, we might want to use the language of existential sustainability also when other values are concerned. In conclusion, it seems that existential risk is a more problematic and less flexible notion than existential sustainability.

Existential threats in psychology

In psychology, an existential threat is a threat to survival [28]. The question “survival of what?” is of consequence [29]. For instance, a person or a group cannot only be defined by their physical existence, but also by their ability to maintain their identity. A person developing dementia continues to exist according to one set of criteria even if the dementia becomes severe, but is at one point no longer the person he or she used to be according to other criteria. There can thus be at least two existential threats involved in such a case. Similarly, a group may experience an existential threat if another group tries to annihilate it, but also when its culture, symbols and beliefs are threatened to the point that the group might transform and change into another unrecognizable entity [29]. In these cases, the number of existential threats might multiply, and if we want to put the concept existential threat to practical use it seems crucial that we clarify what threat we are examining in terms of threats to what. There is a clear parallel to sustainability. The case is similar or worse for sustainability (see e.g. [30]). Borrowing a term from Gallie [31], sustainability might be an “essentially contested concept.” Even when the practical aspects of sustainability discourse are in focus [32], it is usually necessary to first make the sustainability problem clearer by asking questions such as Sustainability–of what?; Sustainability–where?; Sustainability–for whom?; Sustainability–for how long? [33].

The psychological concept of existential threat highlights two things. First, it suggests that “existential” should not be concerned narrowly with physical existence. The potential loss of essential qualities or capacities might also entail existential threat (as, for instance, in [8]). Thus, when Osaka and Yoder’s article “Climate change is a catastrophe. But is it an ‘existential threat’?” [34] continues “So maybe it won’t destroy all humans,” we might justifiably object that it is not the case that a realised “existential threat” entails destruction of humans in that way. It might be the destruction of essential human values, etc., that makes climate change an existential threat. True, that will entail the destruction of human cultures or other human identities. Second, what is noteworthy about the psychologist’s use of “existential” is that an existential threat can be faced by individuals as well as groups of individuals. Hence, it is not necessarily all humans that are threatened when existential threats occur; other significant human units such as individuals, groups or cultures may also be exposed to existential threat.

(It should be noted that “existential threat” is sometimes used outside the psychological context, and may then refer to other kinds of object. For instance: “Mangrove wetlands are facing an existential threat from rapid socio-economic development”, claims [35].

Could we thus suggest that there is a close resemblance between existential sustainability and existential threat?:

X is being existentially sustainable if and only if X is not leading to existential threat

Now, there might be identities that are unimportant for existential sustainability. That is, there might be something about sustainability that entails that only certain identities belong to its domain. For instance, perhaps what is existentially sustainable needs to exist or be operating for a certain time whereas the same does not hold for existential threats. Existential threats might evolve extremely rapidly and what might be threatened can similarly be instantaneous. At least, this potential difference seems worthy to explore.

Moreover, since existential threats might be so specific as concerning only one individual, and perhaps only the identity of that individual, there is a risk that the equivalence above is problematic to uphold for the reason that it becomes vulnerable to complaints that existential sustainability is a “morally repugnant” concept [36], i.e. inviting criticism similar to that of strong readings of sustainability, namely that existential sustainability too easily comes into conflict with other normative values. Some of these problems can be managed by specifying Existential sustainability–of what?; Existential sustainability–where?; Existential sustainability–for whom?; Existential sustainability–for how long? Etc.

Existential dimensions of sustainability–what then are they?

There are as it were then two overlapping but ontologically different versions of existential sustainability, the version where particulars/individuals count and the version where universals/kinds count. It might be prudent to keep these two apart. Furthermore, there are two value-versions existential sustainability as well: one where only “existential” values (that make our lives meaningful) count, and one that can accept other values.

The two ontological versions entail further diversity, especially when the two value-versions are kept apart. Individuals have certain kinds of property and capacity that kinds might lack and vice versa. This means that properties and capacities that make life meaningful might exist at one ontological level but not another. For instance, the existence of a kind implies that there are–and will continue to be–individuals of that kind. Irreversibility exists both on the level of individual and kind, but with the disappearance of a kind, many capacities and properties of individuals of that kind are irreversibly lost. Irreversibility is an important existential dimension of sustainability.

The discussion (summarized in Table 1) entails four versions of existential sustainability:

1A: X does not lead to irreversible loss of unique values on an aggregated level such as a kind, population or cultural level.

1B: X does not lead to irreversible loss of “existential” values on an aggregated level such as a kind, population or cultural level.

2A: X does not lead to irreversible loss of unique values on the level of an individual.

2B: X does not lead to irreversible loss of “existential” values on the level of an individual.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Existential sustainability compared with existential threat and existential risk in three ontological dimensions.

https://doi.org/10.1371/journal.pstr.0000119.t001

1A-2B are structurally similar but do not have the same extension. 2B is arguably the version of existential sustainability articulated in [15]. It does not have an obvious counterpart within the existential threat framework, neither of course in the existential risk discourse. 1A is the primary candidate for an equivalent of Naess’ conception formulated in terms of existential sustainability. 1A has existential risk as a special case, when the aggregate in question is earth intelligent life, but has clearly wider applicability. 1B points to potentially interesting cases where, for instance, cultural values perceived of as “existential” might be at stake. 2A highlights the possibility of unique qualities of individuals, particular qualities that do not exist as universals.

Link to Loss and damage

That existential dimensions of sustainability are tied to irreversibility and existence makes it natural to propose a connection between the existential dimensions of sustainability and loss and damage.

With origins in calls from Small Island Developing States (SIDS) for compensation for climate change impacts, Loss and damage has become a formal part of the United Nations Framework Convention on Climate Change.

The 2015 Paris Agreement (Article 8) recognizes “the importance of averting, minimizing and addressing loss and damage associated with the adverse effects of climate change, including extreme weather events and slow onset events, and the role of sustainable development in reducing the risk of loss and damage.”

Attempts at stipulative working definitions of Loss and damage have been provided, for instance in UNEP:s Loss and Damage: The Role of Ecosystem Services:

“Loss and damage refers to the adverse effects of climate-related stressors on natural and human systems that cannot be, or have not been, avoided through mitigation or managed through adaptation efforts.” [37]

Loss and damage, according to this definition, thus refers to adverse consequences, despite or beyond mitigation and adaptation efforts [38]. Doelle and Seck [39] also contrasts adaptation as being about avoiding the occurrence of harm, whereas loss and damage is about addressing harm that occurs, whether or not it could have been reduced by adaptation.

In other formulations, Loss and damage is about potential consequences, as in UNFCC [40]: “the actual and/or potential manifestation of impacts associated with climate change in developing countries that negatively affect human and natural systems.” Adaptation is about keeping risks within what is perceived as tolerable, whereas loss and damage is a response to risks that cannot be kept within that range [41].

As Boyd and colleagues [42] remark, ambiguous language has arguably been instrumental in order to reach political agreements concerning Loss and damage. As a consequence, Loss and damage probably harbours several distinct concepts, and is understood differently by different stakeholders [42].

The difference between Loss, on one hand, and Damage, on the other, is neither made explicit in the working definition above. However, in several contexts Loss is understood as beyond repair and thus permanent whereas Damage is harm that is potentially reparable or restorable. Moreover, the values affected seem to differ between loss and damage. With their point of departure in UNFCCC:s working definition [40], Thompson and Otto [43] is an example:

“Typically […]“damage” refers to those harms caused by climate change impacts that are capable of being repaired or restored, while “loss” refers to harms caused by climate change impacts, either via extreme events or slow-onset phenomena, that are beyond repair and thus permanent. The kind of losses due to extreme events and slow-onset changes can include loss of human life, livelihoods, culture, customs, and territory, while reparable damages may befall a society’s economy, material infrastructure, or social institutions.” [43]

In connection with loss, human life, livelihoods, culture, customs, and territory are listed in [43]. These exemplify the human condition and typically human aggregates. The damage side in the quote are concerned with economic and social values.

In their survey, Boyd and colleagues [42] identify a dimension of Loss and damage they refer to as “existential”, exemplified by:

“Residual harm, permanent, irreversible, irreplaceable, gone forever, reality, it’s happening, undeniable, unavoidable, nonmarket L&D, non-economic losses, values, sea-level rise, islands, displacement, refugees, loss of homeland, resettlement, reconstruction, rehabilitation, restoration, compensation, ex post, responsibility, anthropogenic climate change, justice, liability, equity, human rights, increase mitigation, more serious about mitigation.” [42]

On any characterization of existential sustainability, if such bearers of value, and values, make up the Loss side of Loss and damage, an overlap between existential sustainability and Loss and damage is implied. If the above observations had been fully generalisable, the following equivalence would have suggested itself:

X is being existentially sustainable if and only if X is not leading to Loss

The imprecision of Loss and damage claimed by [42] would prevent any strict implication relating the two from being true, of course, but this does not hinder the material equivalence from holding under some interpretations of Loss and damage.