What happens when a virus crosses species? Stephan Guttinger looks at viral jumps and the origins of pandemics.

The idea that viruses “jump” is crucial in discussions about epidemics and pandemics. The majority of emerging viral diseases, including SARS, Swine flu, MERS, Zika, and COVID-19, had their origin in animals. The jump to humans thus becomes a central event, both in terms of understanding and intervention. Understanding how pandemics emerge means understanding the conditions under which viruses jump. And localising “the” jump, or places where jumps are likely to happen, provides us with a locus of control, a place where we can interfere and manage the risk of future pandemics.

In the first post of this three-part series, I discussed why viruses should be seen as processes rather than as “things”. In this second post, I want to analyse in more detail why this shift in perspective matters. In particular, I want to discuss how a process-view of viruses can affect our understanding of viral jumps and the emergence of pandemics.

The idea of jumping viruses

At first sight the notion of jumping viruses seems obvious and almost trivial: viruses move from organism to organism. And when they do so, they often don’t touch the ground. Technically speaking, viruses jump.1

But the concept of a “jump” that is used in debates about emerging infectious diseases is richer than such a literal reading of the term. One aspect of this richness is the idea of a border that is being jumped over. The virus is not just doing any jump. It jumps across the boundary of the vessel that contains it (usually a species or an even more abstract entity like “the wild”). Such a border crossing is not an innocent event. It is treated as a “spillover” or an “outbreak”. This not only reflects the idea that viruses are dangerous but also the idea that they have a native resting place from which they should not be dislodged.2

The science writer David Quammen, author of the book “Spillover“, expressed this view in the New York Times as follows:

We invade tropical forests and other wild landscapes, which harbor so many species of animals and plants — and within those creatures, so many unknown viruses. We cut the trees; we kill the animals or cage them and send them to markets. We disrupt ecosystems, and we shake viruses loose from their natural hosts. When that happens, they need a new host. Often, we are it.

We see here the picture of the vessel (“the wild”) that contains things of a particular nature (“unknown viruses”). Humans damage the boundary to the wild and thereby expose themselves to the risk of being jumped on by potentially dangerous entities.

Pandemics, jumps, and a thing-ontology

There are interesting affinities between Quammen’s view and what a thing-view of the world would postulate. In particular, there seems to be a strong assumption of independence and stability at work in the background: the boxes that are postulated, and their contents, are treated as being distinct from each other. There is ‘the wild’ which has its set of properties, and there is humanity (or “civilization”?) that forms a separate realm. Each represents a distinct part of the world that should not be mixed with the other. Viruses rest in their native vessel or “reservoir”. They only become mobile if the conditions are right (elsewhere Quammen compares viruses to dust in an old barn that gets thrown in the air when humans knock down the barn).

This view of viruses has implications for how risk management is handled. If our focus is on the idea of distinct realms that have to be kept apart, the emphasis is on identifying locations where the boundary between them is breached. This is where a dangerous intimacy is created. These “hotspots”, which include locations such as wet markets, now become key sites of intervention that have to be closed or put under surveillance.3

Farming pathogens – a more dynamic picture of viruses

Several authors have criticised the views expressed by Quammen. For instance, Rob Wallace has argued that the way in which Quammen frames the issue ignores the larger causes of the problem, in particular the dynamics and pressures of a global economy that re-shapes local landscapes and cultures, thus fostering the risk of new viral outbreaks.4 Wallace refuses to only focus on “the wild” as a site of risk and opens up the debate to agricultural practices and the business interests driving them. He turns his attention away from wet markets in China and towards the large-scale farming of animals, such as poultry farming. It is in this process, he argues, where new diseases can be bred through the specific practices that are put in place.

Wallace takes the dynamic and context-dependent nature of viruses seriously (without explicitly discussing ontological assumptions), pointing out how changes in animal husbandry can drive the emergence of new and potentially dangerous strains; rather than a singular jump, it is now the way in which animals are kept, circulated, and processed that can lead to the emergence of new pathogens.

Whilst the above article by Wallace focuses on influenza, similar arguments have been made by researchers who study pig-specific viral diseases, such as the Porcine Reproductive and Respiratory Syndrome (PRRS). This syndrome, which costs the swine industry billions of US Dollars every year, only emerged in the late 1980s and quickly spread in swine populations across the globe. Interestingly, this sudden “outbreak” was not triggered by a jump of a new virus into pigs. Rather, it is assumed that the virus responsible for the syndrome has been circulating in pig populations for decades and that it only turned dangerous when intensive farming became common practice in the US and elsewhere. This practice included moving pigs into closed shelters whilst also increasing herd sizes; moving piglets away from their mothers to separate nurseries, either on the same or on a different farm; and introducing the wide-spread use of artificial insemination. All these factors contributed to a new and more dynamic environment that might have allowed the PRRS virus to develop into a more pathogenic form.

Interestingly, the PRRS virus, like corona, influenza, HIV and other RNA-viruses, is thought to operate as a “mutant swarm” or “cloud”, forming the highly dynamic and context-dependent entity that I described in my previous post. What such a mutant cloud is depends on what it does. And what it does depends on its context as much as it depends on “internal” features; in a process view, the notions of “internal” and “external” properties no longer apply in a straightforward manner. This is also echoed by the authors of the PRRS study who argue that “the history of pigs determines the history of PRRS virus”. What makes the virus is the pig.

Processes, viruses, and intimacy

The farming example and its more dynamic view of viruses suggests that Quammen’s view is too narrow. It risks overlooking important avenues of how pandemics can develop, in part because it focuses too much on the idea of a specific “jump”.

At the same time, however, his view might also be too broad. By putting the idea of a jump at the heart of the analysis, any intimacy between humans and “foreign” viruses becomes seen as an exceptional and potentially bad state of affairs. We simply should stay away from any kind of wild animal and the germs they might carry (at some point Quammen recommends that we should leave bats the hell alone).

It is obvious that contact has to be carefully managed because of pathogens such as Ebola or coronavirus. But at the same time, we also have to be careful not to vilify the general flow of viruses too much. For all viruses, activity is fundamental. A virus is not a thing that might become active if the conditions are right. It becomes a thing because it is active. And active viruses, as well as the widespread flow of viruses, can have positive effects, both on individual humans as well as on the human lineage more generally. Viruses can, for instance, help us keep our immune systems primed and ready by creating an immunological imprint. They help us and other organisms defend against pathogenic bacteria. They might reduce the mortality rate in HIV-infected patients. In the longer term, viral flows are also a key force in evolution, constantly adding new sequences to the genomic repertoire of humans and other organisms. Ancient viral insertions, going back millions of years, have been key players in building our immune system. And some even argue that viruses might have helped build the very first cells.5

This of course does not mean that the flow of viruses is always a good thing; it clearly isn’t. But it illustrates that the flow of viruses has to be assessed and managed in a more open way, factoring in potential risk factors as well as potential benefits. Simply disrupting viral flows is not always a good thing. This perspective also aligns with the views of researchers who point to a broader set of factors associated with the risk of future pandemics. It shifts the locus of control away from particular sites, such as wet markets, and towards complex processes and practices that include the flow of capital, animal husbandry, and the dynamics of viral populations.

By Stephan Guttinger


← Part 1


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.



1 — This only applies to horizontal transmission, where viruses are shed by infected cells in the form of virions, which then infect new cells. Some viruses, however, can also integrate their genetic information into the host’s genome. If this happens in germ cells, then the virus is transmitted to the organism’s offspring as an integral part of the host’s DNA.
2 — The repeated use of metaphors here is noteworthy (e.g., ‘jump’, ‘outbreak’, or ‘spillover’). It has been shown that even single-word metaphors can significantly influence how people approach problems and data-gathering. Recently, Brigitte Nerlich discussed metaphors in the context of the current SARS-CoV-2 pandemic.
3 — The idea of hotspots also gained a lot of attention in the early stages of the current SARS-CoV-2 pandemic, with a wet market in Wuhan initially identified as the source of the outbreak. But as researchers gather more data on the virus and its history, they have started to question this theory. Some suggest that a repeated mixing of viruses from an animal source with humans is behind the pandemic. Others have proposed that the virus could have cryptically circulated in humans for several years.
4 — Wallace attacks not only Quammen but also the ‘One Health’ approach that Quammen reports on. The case is complicated as One Health, or at least the views of some of its proponents, is in many ways similar to what Wallace is defending. An analysis of the anatomy of that dispute goes beyond the scope of this post.
5 — It is impossible to discuss all the recent findings on mutualistic viruses in this short blog post. For further details I refer to a paper by Thomas Pradeu, who has delved deeper into the question of symbiotic viruses.