Summary
Many ecocentric moral systems value pristine nature because ecosystems are seen as having interests in survival and proper functioning. However, there are many possible interpretations of what makes an ecosystem better or worse off. Those inclined toward suffering-focused ethical views in the domain of humans and animals may also incline toward giving more ethical weight to the "suffering" of ecosystems than to their "thriving". In this case, it may be best to have fewer ecosystems, exactly contrary to what most ecocentrists assume.
Contents
About this piece
I'm not an expert on environmental ethics, and what I say here may be unoriginal. However, I haven't yet seen someone make this point in the environmental-ethics literature. This might make an interesting topic for a philosophy student to write about further.
Because I haven't read many environmental-ethics writings, I'll focus on just one paper: "The ethics of biological control" by Jeffrey Lockwood. That piece, in turn, relies on the framework developed in Lawrence Johnson's A Morally Deep World.
If someone takes up this project further, the general idea could be applied to biocentric and other ecocentric theories as well.
Interests in survival vs. suffering avoidance
Lockwood explains (p. 4):
Johnson reformulates Aristotle's understanding of Telos (the inherent nature of a being that defines what it is and does) into the concept of well-being interests. To possess interests is to have needs that can be met or, in other words, a condition that can be bettered or worsened in so far as the being is concerned. Johnson argues that some interests are not a reasonable basis for moral action (one could have an interest that was self-destructive, being based on wrong information, addiction, or other such conditions). As such, he narrows the scope of discussion to well-being interests, those elements of an entity's self-identity that allow for the general, effective, and integrated functioning of life processes.
The last sentence reveals a pro-life bias in Johnson's approach. Who's to say that functioning of life processes is most important for well-being? Why not judge avoidance of harm to be more important? I suggest that an ecosystem's well-being interests should primarily be seen as the absence of "suffering", rather than the continuation of life.
More pro-life bias on p. 6:
Johnson asserts that at least some of the interests of species and ecosystems are based on well-being, and therefore these entities are morally considerable. He argues that at the very least, species and ecosystems have survival as a well-being interest. Contemporary definitions of life include some element of self perpetuation; all other biological functions (and morally relevant interests) are impossible if an entity does not survive.
But avoidance of suffering is impossible if an entity does survive. So if suffering reduction takes moral precedence, then nonexistence would seem to be in the well-being interest of ecosystems and species.
What does an ecosystem want?
Lockwood continues (p. 6):
However, it is not just surviving that matters to these entities, but surviving well. That is, species and ecosystems, like any other entities with well-being interests, have an interest in maintaining themselves as coherent, integrated, functioning, ongoing wholes with particular sell-identities. He [Johnson] argues that a species reduced to captive existence in a zoo will deteriorate with respect to its nature and sell-identity. Although Johnson does not pursue this line of reasoning, by logical extension it can be argued that ecosystems restricted to limited, discrete, parcels of land that disconnect them from vital ecological relationships may also lose their identity, their Telos (i.e., island-like preserves may be the zoos of our ecosystems). In this context, Kruess and Tschamtke (1994) have found that fragmentation of the agricultural landscape reduces biological diversity and severely impairs ecological function, particularly the processes of predation and parasitism.
If one objects to zoos on the grounds that animals suffer in them, that's a good argument. But it's less clear that something else of value is lost when an animal is domesticated. Who's to say what's in the interest of the species? Maybe the species was in a negative state previously and now benefits from becoming more domesticated. Likewise, maybe the connected habitat benefitted from being broken up. Maybe predation and parasitism were a scourge that an ecosystem is now freed from. Without a clear theory of species or ecosystem welfare, we can't tell.
Ecosystem responses
When an animal is hungry, it eats food. When an animal is full, it doesn't eat food. In different circumstances, animals react differently, and we generally take these behaviors to be something the animal prefers to do.
When an ecosystem is undisturbed, predator populations are high. When an ecosystem becomes fragmented, populations of large predators decrease. Why not also assume that the loss of large predators is what the ecosystem prefers to have happen in response to a change in circumstances?
Positive feedback
Another way to assess a system's welfare is to see whether it shows reinforcement. For instance, when an animal eats a food and then comes back to eat more of the food, we generally presume that it's in the animal's interest to eat that food. We can generalize this to the principle that when an entity confronts a little bit of X and then later does more of X, X is in the entity's interest.
Now consider a species that gets caught in an extinction vortex. For example:
The R vortex is initiated when there is a disturbance which facilitates a lowering of population size (N) and a corresponding increase in variability (Var(r)). This event can make populations vulnerable to additional disturbances which will lead to further decreases in population size (N) and further increases in variability (Var(r)).
After experiencing a little bit of disturbance, the species responds by getting more and more disturbed. So perhaps it's in the interest of the species to decrease its population size and go extinct?
Likewise, consider a so-called "vicious cycle" in an ecosystem context:
Shrub lands that used to burn every 50 to 150 years on the coast and every 30 to 100 years inland are burning far more frequently in some areas [due to human activity].
The chaparral and coastal sage don't have time to regrow before the next inferno. Hillsides turn into ugly outposts of invasive weeds. [...]
The change in vegetation extends the fire season because many invasive plants burn virtually year-round. Toss a cigarette into a stand of [native] coastal sage in June and in most years it won't ignite. Do the same to a bed of [invasive] red brome and it will.
The ecosystem has responded to an input (increased burning) by setting itself up to burn even more. Hence, presumably the ecosystem prefers to burn more often and thereby reduce species diversity?
One could try to argue that survival is inherently in the interests of a living system. But that's not true for terminally ill people who want to euthanize themselves. Likewise, maybe a feedback loop that results in the loss of an ecosystem is the ecosystem's way of committing suicide.
Buddhism
Another approach to assessing a system's welfare could be to look at whether the system is "restless" and "dissatisfied". Indeed, ecosystems are always changing, suggesting that they're never "content" to stay the way they were. The only way for an ecosystem to escape this suffering seems to be to cease existing.
Conclusion
Obviously, attributing desires to a species or ecosystem feels somewhat strained, and these entities only weakly resemble an agent seeking to improve its condition, which is why I think species and ecosystems deserve a pretty small degree of moral weight relative to animals. But insofar as we do try to care about species and ecosystems, there remains ambiguity about what their interests are and how to weigh interests in survival vs. interests in not continuing to suffer.