How can findings in virology help answer ontological questions of process and substance? In the final post in this series, Stephan Guttinger looks at viral life cycles and the role of intrinsic properties.
In the first two posts of this series, I discussed how a process philosophy can inform our understanding of viruses and pandemics. In this third and last post, I want to look at how virology can inform philosophy, in particular the debate about process and substance ontologies. I will first say more about this philosophical debate and some of the distinctions it is based on. In the second part of the post, I will then turn to virology and to the question of how it can inform philosophy.
Being or becoming – what is more fundamental?
The debate about process and substance ontologies is not only old (reaching back to the writings of Heraclitus and Aristotle) but also complex and often highly technical. It certainly cannot be done justice in a short blog post. However, despite its complexities and age, there are some basic currents that run through this debate and which shape the academic discussion to this day. One of these currents is the question of whether becoming or being represents a fundamental feature of the world. In a substance view, being is usually seen as fundamental. The world is populated by entities that possess intrinsic and unchanging properties. These entities might engage in specific activities, depending on the properties they have. But they don’t have to be active in order to be the kind of thing they are. Activity thus follows from being, and not the other way around.
For a process ontologist, the priorities are set in the exact opposite way: change/becoming is seen as the fundamental feature of the world. What might look like a stable “thing” is in fact a (slow) process. And it is from change that stability follows. This applies most clearly, perhaps, to biological entities such as organisms. As Dan Nicholson and John Dupré put it in the introduction to the essay collection Everything Flows:
As processes, and unlike things or substances, organisms have to undergo constant change to continue to be the entities that they are.
An organism is not like a car that retains its shape and composition, independently of whether the engine is running or not. For an organism to maintain its current appearance, organisation, and behaviour, it has to be metabolically active. But an organism is also constantly changing its appearance and organisation as it goes through its life cycle, developing, for instance, from a zygote into an embryo and eventually into an adult. Moreover, these dynamic features are highly context-sensitive: an organism will develop differently depending on the context it is exposed to (a popular example here is the difference in plant growth depending on the light the plant receives).1 This dynamic and relational nature of organisms illustrates, according to process philosophers, how being follows from activity, rather than the other way around.
Kinds of processes
Viewing the world as consisting of ever-changing entities raises a number of interesting questions. The emphasis on activity and change certainly fits with much of what science and common sense are telling us about the natural world. But even in all this flow there are certain stabilities that characterise organisms and other biological systems. This is especially the case if we abstract away from the nitty-gritty details of metabolic turnover and focus on more general features, such as overall developmental tendencies. At this level, there is more stability than it might seem at first. We know, for instance, that organisms have a “typical” life cycle that moves from one defined stage to the next. The nature of these stages and their sequence are usually specific for each kind of organism. A tiger, for instance, goes through a different set and sequence of stages than a spider or a virus. And even though there might be variation in what the individual stages of a particular life cycle look like, there seems to be a stable core or “program” at the heart of this higher-level process (a tiger will not develop into a spider, no matter how much we tweak its growth conditions). There is something invariant and non-relational that seems to guide the developmental process.
This idea of a program that characterises different kinds of processes has also been used in the literature on process ontology. Nicholas Rescher, for instance, writes:
The basic idea of process involves the unfolding of a characterizing program through determinate stages. The concept of programmatic (rule-conforming) developments is definitive of the idea of process: the unity/identity of a process is the unity/identity of its program.
Such talk of a characteristic program that gives a process its identity is problematic, as it moves a process ontology dangerously close to a substance view. More specifically, it seems to put being in front of becoming: it implies that a process is the kind of process it is because it possesses an internal program that defines its normal unfolding or development.
Unsurprisingly perhaps, this line of reasoning has recently been used to criticise a process view of organisms. Christopher Austin argues that existing process accounts ultimately resort to substance thinking, as they need the idea of fundamental and unchanging properties to make sense of the regularity of life cycle processes.
Dispositions – dynamic or invariant?
The idea that life cycle processes have some sort of program or stable pattern is in itself not a problem for a process view; a process ontology does not deny that some processes are more stable or robust than others. What a process view denies, though, is that stability represents a fundamental feature of the world. The key challenge for a process ontology, then, is not the question of how there can be stable life cycles, but how these stable patterns come about. Are they based on other processes, making organisms processual “all the way down”? Or are they due to intrinsic and unchanging features of biological entities? For Austin the source of biological regularities is clear, at least in the case of organisms: the regularity of their life cycle is explained by a set of developmental “propensities” that each type of organism possesses, independently of everything else. These propensities are grounded in intrinsic dispositions, which form the essence of the kind.2
The idea of dispositions is powerful.3 It allows Austin to account for the dynamic and relational nature of organisms whilst also explaining their stable features. In Austin’s framework the essence of an organism is not a fixed set of structural features (e.g., a particular arrangement of material parts), but rather a dispositional essence or “Bauplan” that allows for variation in developmental outcomes. A disposition can be realised in more than one way, and it can create a number of outcomes, depending on the stimuli its carrier receives. This allows the account to explain the variation that can be observed within and between different kinds of organisms.
Austin’s dispositional view can also explain the regularity of life cycles, because it assumes that dispositions themselves are neither dynamic nor relational; they are the static and intrinsic core that underlies the variation within and between biological kinds. Because of this stable bedrock, biological systems display what could be called a “context-insensitive context-sensitivity”: they undergo constant change and respond to outside inputs, but the way in which they change is unchanging. Each type of organism has its way of developing and of responding to different stimuli, depending on its dispositional essence.
To meet the dispositionalist challenge, the process ontologist will have to find a way of accounting for the regular patterns of behaviour and development that biological systems display, without resorting to the idea of intrinsic and invariant properties. In what follows, I will argue that insights from virology can show us what such an account could look like.
Viral life cycles and the idea of intrinsic properties
The viral world is often depicted in a way that looks much like Austin’s dispositional view. Viruses contain certain parts (proteins) that define what the virus is and what it can do. These parts are inscribed into the viral genome and are key to the unfolding of its life cycle (given the correct enabling conditions), moving from a virion stage to an intracellular phase and, eventually, to the reassembly and the release of new virus particles from the infected cell.
A key event of every viral life cycle is the replication of the viral genome which happens during the intracellular phase; without this stage the cycle could not be completed (as there would be no new viral genomes to package into virions). As we have seen in the first post, a characteristic feature of viruses with RNA genomes is that their replication process is fragile, meaning it has a strong tendency to introduce new mutations. This high mutation rate is functionally relevant, as it leads to extensive sequence variation within the viral population (i.e., the generation of “mutant clouds”). This high level of sequence variation allows the viruses to circumvent existing host immunity and to quickly adapt to new hosts.
The traditional narrative of why RNA viruses display a high mutation rate is very much in keeping with the dispositional perspective adopted by Austin: RNA viruses encode a special molecule, the so-called “RNA-dependent RNA polymerase”, which is required for the production of new viral genomes. This polymerase possesses certain properties that make it “sloppy”, meaning it has a strong tendency to introduce errors. The viral disposition to form mutant clouds is thus defined by the properties of the virus, rather than a set of underlying processes.
However, if we look more closely at the viral replication process this narrative turns out to be too simple. Rather than being determined by intrinsic properties of the virus, the disposition to form mutant clouds is co-produced by a range of processes that run in the integrated virus-cell system. The cell is not just an environment that allows the viral program to unfold its pre-determined activity patterns. The cellular processes are part of what generates and defines these activity patterns in the first place.
We can see the processual underbelly of sloppy replication most clearly perhaps in the case of HIV, an RNA-based virus that displays one of the highest mutation rates of any reproducing biological system.4 Whilst the polymerase that copies the HIV genome is usually described as “error-prone”, researchers analysing the HIV error rate in vivo found that the vast majority of variation is produced by processes driven by the host. Entangled in these processes is a family of enzymes called “apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like 3” (or – to avoid what is probably the worst protein family name in the history of protein biology – “APOBEC3”). The cellular processes involving these enzymes account for up to 98% of the variation observed in replicating HIV genomes.5 The functioning of APOBEC3 enzymes depends on factors both within and outside of the cell, embedding the editing process in a larger set of processes. Part of this set is also the activity of the virus itself, as many viruses have acquired proteins that co-determine the activity of APOBEC3s. HIV, for instance, produces the “Viral infectivity factor” (Vif), a protein that can stimulate the degradation of APOBEC3s. Ultimately, what seems to be the simple consequence of a viral property (the possession of an error-prone polymerase) turns out to be the outcome of a symphony of processes. The viral replication process might be fragile, but this fragility is not an intrinsic feature of the virus (or the cell for that matter).
There is much more that could be said here about the entanglement of viruses and host cells and how this intimate relationship gives birth to a viral life cycle with “characteristic” stages and features. For instance, it has been shown that the activity of APOBEC3 proteins has shaped the composition of human viruses, including HIV and endemic coronaviruses. At the same time, it has been shown that a third of all adaptive changes in conserved proteins in humans, including in APOBEC3s, have been driven by the activity of viruses. What we normally see as two distinct systems that come together in a violent clash (one system fighting the other) are actually different aspects of a deeply entangled network that is not only constantly changing but which also reaches back millions of years. When a “new” virus like SARS-CoV-2 hits the human body, it does not encounter a naïve environment. It rather encounters a molecular landscape that has been shaped and prepared by the shared past of viral-cell systems. The particular life cycle that a current-day virus displays is not based on invariant and intrinsic features of the virus. It is a way of doing things that emerges from an integrated network of processes that shape each other’s shape.
The example of viruses presents us with a narrative that does not focus on the idea of intrinsic dispositions, but which emphasises co-production, dynamicity, and integration. It opens a perspective on organisms and life cycles that prioritises becoming over being, illustrating how virology can challenge and inform the ongoing debate about process and substance ontologies.
Stephan Guttinger is a philosopher of biology based at LSE’s Centre for Philosophy of Natural and Social Science. Apart from looking at viruses and process ontology, he is also interested in how biologists create trustworthy knowledge, and how conceptual changes in the life sciences affect broader debates about science and health policy.
Featured image: Circle Vectors by Vecteezy