On 2014 Oct 09, Arturo Casadevall commented:

We thank Joshua L. Cherry for the additional comments. We also feel that our original response applies to much of the new comments but will try again to clarify our answers. As we have done previously, we copy Joshua L. Cherry’s comments and respond below.

Joshua L. Cherry states “We have all encountered writing that refers to science or its subfields and uses scientific language but lacks scientific content. The editorial does something similar with respect to philosophy. It defines epistemology and makes use of its vocabulary, but the arguments that it makes are not epistemological. The only general principle about knowledge that it invokes is the Faberite notion that knowledge is good. This is not an epistemological proposition or a philosophical insight, nor a revelation to scientists”.

We feel that Joshua L. Cherry misses the point in continuing to argue that our essay is not about philosophy and its branch epistemology. GOF experiments are the gold standard for acquiring certain times of information including the capacity of a virus to become mammalian transmissible. GOF experiments are accepted methodology in the field and commenting on their power and their standing within the normative standards of the field of molecular microbiology clearly constitutes epistemological content for scientific methodology is part of the epistemology of science. These experiments provide information with the highest standard of rigor that is unobtainable in any other way.

Joshua L. Cherry states that “The benefits of GOF experiments commonly discussed are indeed benefits of the resulting knowledge. (What else would they be? The economic benefits of creating employment for laboratory personnel?) These have mostly been supposed direct practical applications of this knowledge for preventing human H5N1 infections. The response above suggests that the authors intended to emphasize “the many other possible future uses of that knowledge, some of them practical, but some of them purely theoretical”. As I stated, most or all scientific knowledge has many possible future uses, and it is common knowledge, not a philosophical insight, that science works this way.”

We have no disagreement with this paragraph although we note that if this is indeed ‘common knowledge’ then we have trouble understanding why it is necessary for us to point this out in the context of weighing the epistemic value of GOF experiments.

Joshua L. Cherry states that “Exceptional risks demand exceptional benefits. Satisfaction of normative standards does not imply exceptional benefits. The authors indeed make a logical leap in concluding that GOF experiments must be powerful because they share certain formal properties with experiments that proved to be powerful. The literature is filled with results of experiments that meet these normative standards, with importance ranging from great to nearly nil. Thus, we must consider the value of these particular experiments, using scientific reasoning and judgment. That is exactly what most of the debate has been concerned with.”

We see several problems with this paragraph. First, there is no evidence that the experiments done carry ‘exceptional risks’. These have been done in laboratories with rigorous biosafety protocols and high level bio-containment capabilities. Furthermore, we do not know whether transmissibility in Ferrets confers transmissibility in humans. Second, we never argued that GOF experiments yielded "exceptional benefits," alone capable of balancing exceptional risks. Instead, we merely argued that the epistemological value of GOF experiments must be part of the bookkeeping, and we questioned the objectivity of the claimed exception character of the risk. These experiments have already provided important information regarding the biological potential of highly pathogenic avian influenza virus to acquire the capacity for mammalian transmissibility. We did not assign a specific value to this epistemic gain: we simply argued that it needed to be part of the risk-benefit analysis. We have no problem with the comment that “we must consider the value of these particular experiments, using scientific reasoning and judgment” and if Joshua L. Cherry feels that this “is exactly what most of the debate has been concerned with” then we are mostly on the same page.

On 2014 Oct 02, Joshua L Cherry commented:

I thank the author(s) for responding. My original comment applies to much of the response, but a few points should be clarified.

We have all encountered writing that refers to science or its subfields and uses scientific language but lacks scientific content. The editorial does something similar with respect to philosophy. It defines epistemology and makes use of its vocabulary, but the arguments that it makes are not epistemological. The only general principle about knowledge that it invokes is the Faberite notion that knowledge is good. This is not an epistemological proposition or a philosophical insight, nor a revelation to scientists.

The benefits of GOF experiments commonly discussed are indeed benefits of the resulting knowledge. (What else would they be? The economic benefits of creating employment for laboratory personnel?) These have mostly been supposed direct practical applications of this knowledge for preventing human H5N1 infections. The response above suggests that the authors intended to emphasize “the many other possible future uses of that knowledge, some of them practical, but some of them purely theoretical”. As I stated, most or all scientific knowledge has many possible future uses, and it is common knowledge, not a philosophical insight, that science works this way.

Exceptional risks demand exceptional benefits. Satisfaction of normative standards does not imply exceptional benefits. The authors indeed make a logical leap in concluding that GOF experiments must be powerful because they share certain formal properties with experiments that proved to be powerful. The literature is filled with results of experiments that meet these normative standards, with importance ranging from great to nearly nil. Thus, we must consider the value of these particular experiments, using scientific reasoning and judgment. That is exactly what most of the debate has been concerned with.

On 2014 Oct 01, Arturo Casadevall commented:

We thank the writer for his comments. The writer is critical of several of the statements in our editorial. We respond to each individually:

Joshua L Cherry states that: ‘The arguments are framed in philosophical terms, but there is no philosophical content, and the underlying question is one of scientific judgment.’

We found this comment puzzling. Our paper includes discussion on two of the branches of philosophy: epistemology and ethics (by the writer’s own admission). Framing the GOF experiments within the normative standards of the field microbiology explores how we seek knowledge in the area. This together with discussion of ethical issues constitutes philosophical content.

Joshua L Cherry states that: “[W]hen one does a risk-benefit analysis of this issue, the epistemic gain from GOF experiments should be included in the bookkeeping: if one does that, the benefits of GOF experiments are potentially so great as to warrant our risking more than we otherwise might.” This is a puzzling assertion. All of the benefits that have been considered are benefits of the resulting knowledge (“epistemic gain”). Perhaps the authors mean to refer to an intrinsic good of having knowledge, independent of any practical applications. If so, it is possible, and helpful, to state this more plainly, as I have just done. If there is a philosophical issue here, it is one of ethics, not epistemology. Perhaps the authors are referring to the practical value of the knowledge beyond its direct applicability to the case at hand. In either case the argument is weak. These are common properties of experiments, not special properties of potentially dangerous GOF experiments. The resources used for such experiments could be reallocated to other experiments that lack the serious safety concerns yet also have epistemic value (on any meaning), and quite possibly more of it.’

The writer states that ‘all the benefits that have been considered are benefits of the resulting knowledge’. This is not the case. A reading of our paragraph describing the knowledge gained from GOF experiments considers the information that highly pathogenic avian influenza viruses can become mammalian transmissible useful because it provides a warning that these viruses have the biological potential to acquire this property. Prior to these experiments there was debate as to whether these viruses could become transmissible. Now we know that they can and that is a clear epistemic gain. But that epistemic gain goes beyond the utilitarian gain from any immediate application of this new knowledge, in part because of the many other possible future uses of that knowledge, some of them practical, but some of them purely theoretical. We have no disagreement with the notion that the subject of ethics is interwoven into considerations for GOF experiments. However, the idea that we should reallocate resources away from this type of work would accomplish nothing in helping us prepare against an influenza pandemic. As the currently developing Ebola epidemic in West Africa shows us we must continue research into highly dangerous pathogens for it is knowledge in medicine and science that provide the defenses against new threats from nature.

Joshua L. Cherry states that: ‘The authors make much of the claim that the GOF experiments meet the “normative standards of the fields of microbiology and infectious diseases”. This is faint praise for experiments that, many fear, might cause pandemics. We might question the merits of these standards, despite their being normative, and might ask whether they are sufficient conditions for value, or merely necessary. Even putting those questions aside, at best this point establishes that the experiments have nonzero value, not that they are exceptionally powerful, as must be the case to justify exceptional danger. Again, there are plenty of other experiments that also satisfy these standards without posing the same threat.’

Meeting the normative standards of the fields of infectious diseases and microbiology is high praise for GOF experiments for these fields have delivered some of the greatest successes in medicine. The writer appears to be mixing two issues that are very different: the fear and the power of GOF experiments. Fear can be minimized with strict laboratory safeguards that greatly reduce the likelihood of accidents. With regards to the power of GOF experiments, these have already taught us that highly pathogenic avian influenza virus has the capacity for mammalian transmission, as noted above. Since mammalian transmission is necessary for a pandemic this information is incredibly important. Humanity now stands warned that with a few sequence changes H5N1 influenza virus can acquire the property of mammalian transmissibility. The writer states that ‘there are plenty of other experiments that also satisfy these demands without posing the same threat’ but provides no examples. In fact, we counter that there is no other technology or mechanism short of waiting for epidemic to occur that could have provided this information. If there were, it is likely that a scientist would have pursued it by now. Hence, we stand by the words that these experiments are exceptionally powerful and that we lack alternatives if we are to seek certain types of knowledge.

Joshua L. Cherry states that: The editorial seems to suggest such fallacious reasoning as this: The discovery that HIV causes AIDS established a cause-and-effect relationship about a pathogen and was very powerful; GOF experiments have established cause-and-effect relationships concerning pathogens; therefore, GOF experiments are very powerful.

The key word here is “seems”: it is unclear how the writer infers this from what we wrote. GOF experiments trace their ancestry to the powerful scientific techniques that that have been responsible for some of the greatest successes in medicine including establishing disease causation and intervening against such diseases with vaccines and drugs. GOF experiments are fully within the normative standards of microbiology and molecular biology and as such they are incredibly powerful investigative tools that humanity can use to learn about microbial threats. As noted above GOF experiments have already produced highly valuable information that is not available from any other source.

On 2014 Sep 23, Joshua L Cherry commented:

This editorial purports to offer new considerations that favor the continuation of potentially dangerous “gain of function” (GOF) experiments with pathogens with pandemic potential. The arguments are framed in philosophical terms, but there is no philosophical content, and the underlying question is one of scientific judgment. The philosophical language serves only to obscure and to create the illusion that a new consideration is being introduced.

The authors state that

“[W]hen one does a risk-benefit analysis of this issue, the epistemic gain from GOF experiments should be included in the bookkeeping: if one does that, the benefits of GOF experiments are potentially so great as to warrant our risking more than we otherwise might.”

This is a puzzling assertion. All of the benefits that have been considered are benefits of the resulting knowledge (“epistemic gain”). Perhaps the authors mean to refer to an intrinsic good of having knowledge, independent of any practical applications. If so, it is possible, and helpful, to state this more plainly, as I have just done. If there is a philosophical issue here, it is one of ethics, not epistemology. Perhaps the authors are referring to the practical value of the knowledge beyond its direct applicability to the case at hand. In either case the argument is weak. These are common properties of experiments, not special properties of potentially dangerous GOF experiments. The resources used for such experiments could be reallocated to other experiments that lack the serious safety concerns yet also have epistemic value (on any meaning), and quite possibly more of it.

The authors make much of the claim that the GOF experiments meet the “normative standards of the fields of microbiology and infectious diseases”. This is faint praise for experiments that, many fear, might cause pandemics. We might question the merits of these standards, despite their being normative, and might ask whether they are sufficient conditions for value, or merely necessary. Even putting those question aside, at best this point establishes that the experiments have nonzero value, not that they are exceptionally powerful, as must be the case to justify exceptional danger. Again, there are plenty of other experiments that also satisfy these standards without posing the same threat.

Through reference to “normative standards”, the authors make an unjustified connection between GOF experiments and such important findings as the discovery that HIV causes AIDS. They tell us that “GOF experiments are very powerful because such experiments can give direct information on cause-and-effect relationships” about infectious agents. In reality, not all experiments that provide such information are “very powerful”. The editorial seems to suggest such fallacious reasoning as this: The discovery that HIV causes AIDS established a cause-and-effect relationship about a pathogen and was very powerful; GOF experiments have established cause-and-effect relationships concerning pathogens; therefore, GOF experiments are very powerful.

Debate about GOF experiments has always been about whether the value of the knowledge they might yield is sufficient to justify their risks. Many claims for particular benefits have been abandoned, and some have been undermined by arguments put forth to downplay safety concerns. Referring to knowledge as “epistemic gain” adds nothing to the debate.

On 2014 Sep 23, Joshua L Cherry commented:

This editorial purports to offer new considerations that favor the continuation of potentially dangerous “gain of function” (GOF) experiments with pathogens with pandemic potential. The arguments are framed in philosophical terms, but there is no philosophical content, and the underlying question is one of scientific judgment. The philosophical language serves only to obscure and to create the illusion that a new consideration is being introduced.

The authors state that

“[W]hen one does a risk-benefit analysis of this issue, the epistemic gain from GOF experiments should be included in the bookkeeping: if one does that, the benefits of GOF experiments are potentially so great as to warrant our risking more than we otherwise might.”

This is a puzzling assertion. All of the benefits that have been considered are benefits of the resulting knowledge (“epistemic gain”). Perhaps the authors mean to refer to an intrinsic good of having knowledge, independent of any practical applications. If so, it is possible, and helpful, to state this more plainly, as I have just done. If there is a philosophical issue here, it is one of ethics, not epistemology. Perhaps the authors are referring to the practical value of the knowledge beyond its direct applicability to the case at hand. In either case the argument is weak. These are common properties of experiments, not special properties of potentially dangerous GOF experiments. The resources used for such experiments could be reallocated to other experiments that lack the serious safety concerns yet also have epistemic value (on any meaning), and quite possibly more of it.

The authors make much of the claim that the GOF experiments meet the “normative standards of the fields of microbiology and infectious diseases”. This is faint praise for experiments that, many fear, might cause pandemics. We might question the merits of these standards, despite their being normative, and might ask whether they are sufficient conditions for value, or merely necessary. Even putting those question aside, at best this point establishes that the experiments have nonzero value, not that they are exceptionally powerful, as must be the case to justify exceptional danger. Again, there are plenty of other experiments that also satisfy these standards without posing the same threat.

Through reference to “normative standards”, the authors make an unjustified connection between GOF experiments and such important findings as the discovery that HIV causes AIDS. They tell us that “GOF experiments are very powerful because such experiments can give direct information on cause-and-effect relationships” about infectious agents. In reality, not all experiments that provide such information are “very powerful”. The editorial seems to suggest such fallacious reasoning as this: The discovery that HIV causes AIDS established a cause-and-effect relationship about a pathogen and was very powerful; GOF experiments have established cause-and-effect relationships concerning pathogens; therefore, GOF experiments are very powerful.

Debate about GOF experiments has always been about whether the value of the knowledge they might yield is sufficient to justify their risks. Many claims for particular benefits have been abandoned, and some have been undermined by arguments put forth to downplay safety concerns. Referring to knowledge as “epistemic gain” adds nothing to the debate.