5 / 06
Image of birds flying. Image of birds flying.

#355 Animal Pain Re-visited

February 02, 2014

Dr. Craig comments:

I have personally found Michael Murray’s work on the problem of animal pain—or perhaps, better, animal suffering—to be both thought-provoking and helpful. After I cited his work in debate, several persons responded to his arguments, prompting Reasonable Faith readers to send in questions about the same. Here Dr. Murray, at my invitation, responds to these critics. What I find interesting about his response below is that he refers only obliquely to what I take to be one of his strongest points: that non-human animals do not have a first-person perspective on their experiences, including experiences of pain, that is to say, they cannot adjoin to their experiences the prefix “I think/feel that. . . ,”so that animals, even if in pain, are not aware that they are themselves in pain. Wholly independently of this point, Dr. Murray shows that his critics have not shown that animal suffering of a morally significant sort really exists.—William Lane Craig

Flag of United States. United States

Photo of Dr. Craig.

Dr. craig’s response


Michael Murray responds:

Throughout my philosophical career I have sought to address what I take to be some of the most difficult intellectual challenges to the Christian faith: the hiddenness of God, naturalistic accounts of the origin of religious belief, the reality of hell, etc. In a 2008 Oxford University Press book, Nature Red in Tooth and Claw, I addressed another of these challenges: the problem of evil as it is manifested through the pain, suffering, and death of non-human animals (henceforth, “animals”), especially as that is understood within current evolutionary theory. Since the time of Darwin, critics of Christianity argued that the extent, duration, intensity, and quantity of evil entailed by Darwinism provided powerful evidence against the existence of God. The book is aimed at assessing whether or not these phenomena indeed provide such evidence.

The book analyzes a variety of different ways that Christians might respond to the problem of animal suffering and assesses them individually. The goal was not to provide a single “response” to the “problem of animal suffering” but to look at a variety of responses and assess their strengths and weaknesses.

I am pleased to say that the book has received a good deal of attention. Like any good philosophical book, it contains arguments that are controversial and provocative. However, a great deal of attention has been focused on one small part of one argument that I discuss in chapter 2—an argument that I will revisit here. The argument has been criticized by some professional philosophers in print. However, it has also become something of a cause célèbre after it was referenced by Bill Craig in some of his public debates. Critics of Craig’s use of the argument have claimed that the argument is scientifically uninformed and some have gone so far as to request that Bill apologize for using the argument (!). The argument has now become the subject of at least two YouTube “exposés” which have been linked to and reposted by dozens of sites.

So what has caused all of the fuss?!? Before I get to that, let me take a few words to situate the argument overall. The aim of the book is to undermine an argument that the reality of animal suffering is good reason to reject the existence of God. Like most “arguments from evil” the claim is that some evil exists, and that there is no morally sufficient reason for God to permit that evil to exist. If there were a God, that evil would thus not exist. But it does. So there is no God.

To resist that argument you need to show either (i) that the evil does not in fact exist (or that we are not in a good position to affirm that it does exist), or (ii) that there is a morally sufficient reason for God to permit the evil (or that we are not in a good position to tell whether or not there is a morally sufficient reason for God to permit the evil). In the book I try to resist the argument by all of these means.

However, in Chapter 2 I look specifically at arguments along the lines of (i). That is, I look at arguments that hold that animal pain and suffering are not real, or that we are not in a good position to know whether or not they are real.

Needless to say, this argument is not popular. Anyone who has watched nature films, or been to a farm or zoo, or has a pet, thinks it is clear and obvious that animals feel pain and experience suffering. What sensible person would deny this? Well, maybe none! But philosophers are always willing to explore counter-intuitive positions to see whether or not they have any merit. And in this chapter I do just that.

The arguments of the chapter are complicated and I cannot reiterate them here in detail. But in brief, I consider the following question: how confident are we entitled to be that animals experience the sort of pain and suffering that counts as evil? That might sound like a weird way of putting it, since one might think that all forms of pain and suffering are evil. And that is where things get complicated.

Organisms encounter all sorts of dangers in their environments. Some of these are stimuli that directly threaten the organism’s biological integrity and well-being in some way. Pain theorists call these “noxious stimuli.” Animals of all types react to noxious stimuli, and in a variety of ways. In most cases, the reaction involves some type of avoidance behavior that aims to get the organism away from danger and back to safety.

Sometimes these noxious stimuli involve pain and sometimes they don't. When I withdraw my hand from the hot stove, there is pain. When the snail withdraws into its shell when touched (arguably) it does not (it simply lacks the necessary neural complexity). So the question is: when do noxious stimuli and aversive behaviors involve feeling pain and when do they not? That turns out to be a very difficult question. The reason it is difficult is that answering that question requires that we know, not whether animals can react to noxious stimuli (of course they do that!), but whether they can “feel the pain” that is (sometimes) caused by noxious stimuli.

Answering this question requires that we have some theory about what it means to be able to “feel the pain” or to “feel” more generally. As it turns out, this is one of the most difficult questions in philosophy of mind and, indeed, in all of contemporary philosophy. “Feeling” of this sort is known as “phenomenal consciousness.”

In Chapter 2 of my book I look at a variety of ways of thinking about phenomenal consciousness and its implications for the problem of animal suffering. One theory I consider in some detail is known as the “Higher Order Thought” (or HOT) theory of consciousness. On this theory, we can distinguish between “first order” mental states and “higher order” mental states, and it is higher order states that are required for “feeling” or “phenomenal consciousness.” There is actual evidence for these different orders of mental states in phenomena such as blindsight, where patients with a certain sort of brain damage report that they are blind, and yet they are able to navigate obstacles in their path by using their eyes. How do we explain that sort of thing? The answer is: blindsight patients have access to the existence of the obstacles through visual means, but they are not aware that they have this access. So they are conscious of the objects in one sense (sometimes called “access consciousness”) but not in another sense, (“phenomenal consciousness”). And HOT theories provide a plausible explanation of how this could be so.

Thus when we see animals reacting in pain, it is a fair question to ask: do they have phenomenal consciousness of the pain, or just access consciousness of the noxious stimulus? If HOT theories are right, they can only have phenomenal consciousness if they have this higher order thought or awareness of the first order mental states that are caused by the noxious stimulus. Do they have such higher order thoughts? Well, that is controversial. It is just hard to know if animals have higher order thoughts, and hard to figure out what sort of behavioral or neural evidence would confirm this. So if we are not sure if animals have phenomenal consciousness of pain, and that is the sort that matters when it comes to pain counting as bad or evil, then we are not sure if there is a “problem of animal suffering.” Call this “response 1” to the problem of animal suffering.

Arguments of this general kind are not really controversial from the standpoint of contemporary philosophy of mind, despite the fact that most of us are intuitively convinced that animals feel pain. But interestingly, the argument above is not the one that has caused all of the fuss on YouTube and the web. To see the locus of that controversy we need to turn to the next argument.

Later in that same chapter I consider another strange possibility when it comes to animal cognition that goes like this. Let’s imagine that animals have first order mental states, and have higher order thoughts concerning those states, and thus experience phenomenally conscious pain. Does that show that animals suffer? Not exactly. The reason for that is as follows. When we have phenomenally conscious states, those states can feel good (tasting chocolate), or feel bad (smelling sulfur), or neither good nor bad (seeing red). Is it possible that animals have phenomenal conscious awareness of pain, but that awareness does not feel bad? Well, that would seem to be ridiculous! How could animals feel pain and that pain not feel bad!?!?

As strange as it might be, however, we know that in some cases phenomenal consciousness of pain does not in fact feel bad. During the 1940s and ’50s the surgical procedure that we now call “frontal lobotomy” was scarily common. During the very crude surgical procedure surgeons removed portions of the prefrontal cortex of patients, with some surprising results. In some cases the procedure was carried out to resolve chronic pain. While it did not always work, there were some cases where something very weird happened—patients reported that they still felt the pain but that it “did not bother them” any longer.

Because of cases like this, I asked the following question: what if animal consciousness of pain were like that? That is, what if, when it comes to animal pain, the corresponding mental state does not have a negative valence to it? That is, what if they experience pain the way these lobotomy patients did? Would there be a “problem of animal pain”? I argue that the answer is no. If that were to happen, animals would “feel pain” but they would not “suffer” from it. Call this “response 2” to the problem of animal suffering.

Do we know that any of the possibilities I describe (that animal pain is like blindsight—experienced but not felt (response 1)or like “lobotomy pain”—felt but not undesirable (response 2)) are true in the case of animals? Absolutely not. But do we know that they are not the case? I argue that we don’t. That is, for all we know, animal pain works as described in one of those options. The scientific data do not resolve the philosophical question. And thus we don’t know if we have a problem of animal pain.

Do people disagree about this? Yes. However, I think that most scientists who think about these questions do not have the full range of distinctions that philosophers of mind do when reasoning about these matters. So I think that the scientists who disagree don’t fully appreciate the relevant philosophical subtleties. (However, some neuroscientist/philosopher collaborations have been looking at exactly such issues, indeed suggesting that the prefrontal cortex plays just the sort of role I suggest when it comes to phenomenal consciousness—whether of pain or other sensory states. See for example, Lau and Rosenthal, “Empirical Support for higher-order theories of conscious awareness,” Trends in Cognitive Science, August 2011, vol. 15, no 8.)

But what caused all of the online controversy was this. When discussing “response 2,” I noted that when the prefrontal cortex (PFC) was disrupted, the negative feelings that go along with having pain were disrupted as well. Since this part of the mammalian brain was last to evolve, perhaps most animals do not experience negative-feeling-pain (because they lack a PFC).

This became the point of controversy. YouTube channels like skydivephil toted their video cameras around to interview a bunch of scientists and asked them whether or not animals feel pain (an exercise not likely to address the philosophical subtleties of the arguments I address), and have PFC’s. All of those interviewed indicated that other animals have PFC’s, and thus my argument fails.

Well, there are three things to say about this. First, even if that were right, my argument does not fail. What I claim is that, for all we know, animal pain occurs without “negative-feeling.” Even if the absence of PFCs in humans often correlates with pain-without-negative-feeling, that does not mean that organisms that have PFCs experience negative-feeling-pain. Why? Because we are not sure if animal cognition is subserved by the same structures as they are in human beings. As a result, even if they have PFCs, that does not guarantee that they have negative-feeling-pain.

Second, it is not obviously correct that animals outside of humans and higher primates have PFCs. As even the folks at skydivephil note, there have been different ways of demarcating the PFC over time. For those not familiar with neuroanatomy it is worth pointing out that identifying regions of the brain is not like opening up the abdomen and distinguishing the stomach and kidney and liver. Brain regions are contiguous with each other, and there are different criteria that can be used to discriminate between regions. In the early twentieth century, the PFC was demarcated by location and cell type. Humans and higher primates have a certain cell type (known as “granular”) that composes a specific cortical layer, and the PFC was identified with this layer. Some later anatomists discarded this criterion for demarcating the PFC, in part because it made it hard to find a PFC in non-higher-primates. Thus, later anatomists defined the PFC functionally as the projection zone from another part of the brain known as the thalamus. Some non-primates do have such projection zones.

Is scientific opinion now unanimous on how to think about the PFC? No, it is not. Indeed some recent critics argue directly against the functional criteria that entail that non-primates have a PFC. (See H. J. Markowitsch and M. Pritzel, "The prefrontal cortex: Projection area of the thalamic mediodorsal nucleus?"Physiological Psychology 7 (1) (1979): 1–6; and

T. M. Preuss, "Do rats have prefrontal cortex? The Rose-Woolsey-Akert program reconsidered," Journal of Cognitive Neuroscience 7 (1) (1995): 1–24). So the jury is not fully in on this question.

In any case, we can ask the following question: what brain structure is important here for negative-feeling-pain? A granular cortical layer IV or a projection zone from the thalamus? We don’t know. But we do know when “that area” in humans is disrupted, then negative-feeling-pain sometimes disappears. So if animals lack what we have in this part of the brain, then perhaps they lack this sort of pain as well. That conclusion remains intact.

Finally, even if non-primates have PFCs, the human PFC is completely different from every other type of organism. Indeed one recent survey of primate neuroanatomy describes the human PFC as “absolutely, obviously, and tremendously” different (Rilling, Trends in Cognitive Science, vol. 18, no.1 (January 2014)). If those differences (which are destroyed in a lobotomy) are what makes negative-feeling-pain possible, then perhaps animals do not have such pain.

Let me remind everyone that this controversy concerns only one small claim that concerns just one of the arguments of this chapter. Even if we throw all of “response 2” away, there are other arguments that might be used to undermine the argument from animal suffering by showing that animals don’t experience phenomenally conscious and morally relevant suffering. But given the criticisms that have been leveled by the online objectors, we don’t yet have reason to abandon “response 2.”

- William Lane Craig