What does it take for your life to go better or worse? One idea is experientialism. For experientialists, what matters is sentience, the capacity to experience pain and pleasure. Experientialists typically appeal to a distinction between moral agency and moral patiency to argue that only sentient beings can be moral patients. The paradigm moral agent is the adult human, capable of both thinking morally and acting morally. Most moral agents are also moral patients because most adult humans are sentient. The paradigm moral patient that is not also a moral agent is a newborn baby or a nonhuman animal. For my purposes, the key doctrine of experientialism is that sentience is necessary for both moral agency and moral patiency.
The goal of this paper is to refute that doctrine and argue that the capacity for reflection by itself is sufficient for both moral agency and moral patiency. In other words, a purely reflective but insentient being would be both a moral agent and a moral patient simply in virtue of their capacity for reflection. Who explicitly denies this? Suchy-Dicey (2009) argues that a being that was reflective but not sentient would not be a moral patient. She states that “autonomy without the potential for experiencing welfare is not valuable…the ability to experience welfare is a precondition for the value of autonomy” (2009, p. 134). Thus, Suchy-Dicey says the value of reflection is parasitic upon sentience but not vice versa. That is, an entity is a moral patient if it is both sentient and reflective, or if it is only sentient—but if an entity is reflective but not sentient then on Suchy-Dicey’s view it does not count as a moral patient. Hence, Suchy-Dicey’s view is characterized by two features:
(1). Value Pluralism: Both sentience and reflection are intrinsically valuable.
(2). Value Asymmetry: The value of sentience for moral patiency is independent of reflection but the value of reflection for moral patiency is dependent on sentience. Thus, if an entity is reflective but not sentient, it is not a moral patient.
I agree with (1) but deny (2). Instead, I will defend the following thesis:
(2*). Value Symmetry: the value of sentience for moral patiency is independent of reflection and vice versa. Thus, an entity that is reflective but not sentient would still be a moral patient.
This paper aims to defend (2*) against (2). To do so, I defend the following argument:
- Experientialism assumes that all moral patients and all moral agents are necessarily sentient.
- The capacity for reflection by itself is sufficient for both moral patiency and moral agency.
- By (2), if a purely reflective being existed, it would be both a moral patient and a moral agent.
- Purely reflective beings can exist.
- Thus, experientialism is false.
Premise (1) just falls out of the commitments of experientialism. The most controversial premise is arguably (2). To defend it, I will need to do several things. In section 2, I will explain what I mean by “the capacity for reflection”, explain why it’s sufficient for moral agency, and argue that purely reflective beings can exist. In section 3, I will continue by arguing that reflection is sufficient for moral patiency. Doing so will provide the needed ammunition to argue against experientialism.
2. What is reflection?
The paradigm reflective agent is a normal human adult, capable of reflective self-consciousness. Gallagher’s (2010) definition of reflective self-consciousness is a good place to start. He defines it as “an explicit, conceptual, and objectifying awareness that takes a lower-order consciousness as its attentional theme.” Several themes are important for my understanding of reflection. First, it must be explicit. A cat might think “I am hungry” but this thought is never explicitly articulated in its mind in the way a reflective human might reflect to themselves, “Boy if I don’t eat breakfast I’m going to be hungry this evening for sure.” Second, reflection must be conceptual. What I mean by that is that in order to reflect one must have the concept of “reflection”, or at least some concept of “consciousness”. A cat might have a psyche but it lacks a concept of psyche qua psyche. A reflective creature knows as its reflecting that it’s reflecting because it has at least one concept about reflection as such to distinguish it from other psychological events like behaving or perceiving.
Thus, to reflect in the full sense I intend one must have an explicit understanding of what it means to reflect and the ability to know that you are reflecting when you are reflecting. Furthermore, a distinguishing feature of reflection is that a reflective creature can reflect on just about anything: themselves, trees, rocks, numbers, philosophy, art, reflection itself, evolution, space-time, etc. While there might be some contents that are too unwieldy for human reflective agents to fully reflect on, a defining feature of reflection is its flexibility with regard to the contents of reflective acts. If a reflective agent is relaxed and not pressed for time it can very well reflect on almost anything so long as it has the right conceptual repertoire. Thus, I avoid the term “reflective self-consciousness” because reflective agents can actually take as an object of reflection just about any object or proposition, not just the “self”. Hence, I prefer to talk about “reflective consciousness” i.e. reflection. A feature of reflection closely related to flexibility is the ability to switch between different objects of reflection. A reflective creature, when suitably relaxed, can choose what to reflect on when it wants to. If it wants to reflect on the past, it can; if it wants to reflect on the future, it can.
Phenomenologically speaking, reflection is spatial, selective, and perspectival. Reflection is spatial because if I asked you to reflect on your cat and then your dog you would not imagine them mushed together; you would first reflect on your cat and then “move” onto your dog. All reflection is spatialized in this sense because the objects of reflection are “separated” from each other in mental space. This applies to the most abstract of ideas: if I ask you to reflect on the concept of liberty and then reflect on democracy there will be “movement” in your act of reflection as you go from idea to idea. Reflection is selective because if I reflect on what I had for breakfast yesterday, I cannot simultaneously reflect on what I want for breakfast tomorrow. Reflection is perspectival because if I reflect on my walk through town yesterday the reflective act is done from a perspective. If my reflection is veridical I might reflect as if I were peering out of my head bobbing up and down as I walk but in all likelihood my reflection will be disembodied like a camera floating freely through space able to fly through the city at any speed.
Another feature of reflection is the capacity to explicitly reason and articulate about intentional actions qua intentional actions. To interact with something nonreflectively is to interact it without explicitly realizing you have done so and without the ability to give a reason why you have done so. Conversely, to interact with something reflectively enables you to reflect on your reasons for having chosen the action you did and the ability, if needed, to explicitly articulate your reasons for having acted in the way you did. The reasons you give might not be indicative of the true, underlying causal mechanisms for your action but what’s important is the ability to articulate in terms of intentional actions even if you are confabulating (Nisbett & Wilson, 1977). Moreover, even if your voicebox or muscles were completely paralyzed you would still have the ability to articulate your reasons so long as you can articulate them to yourself or so long as you possess the knowledge that if you had a means of expressing yourself you could actually articulate. Thus, what counts is not so much the literal articulation of reasons but the capacity or potential to articulate reasons for action. Moreover, by action I mean mental or behavioral action e.g. you could articulate to yourself why you chose to imagine yourself playing tennis as opposed to imagining yourself walking through your house.
Now that I have explained part of what it means to be a reflective agent, I want to explain why reflective agents are also moral agents, what I call reflective moral agents. Defending the cogency of reflective moral agency will clear the ground for my defense in the next section of reflective moral patiency. It’s relatively uncontroversial the ability to reflect has instrumental value for moral agents insofar as reflective creatures could reflect on better ways to help moral patients but why should reflective agents be moral agents just in virtue of their being reflective agents and not because reflection is instrumentally valuable? One reason is that reflective agency is important for realizing many things of intrinsic value according to what has been called “objective list” approaches to intrinsic goodness. Common items on these lists of intrinsically valuable goods include things such as: developing one’s talents, knowledge, accomplishment, autonomy, understanding, enjoyment, health, pleasure, friendship, self-respect, virtue, etc. Arguably reflection is not crucial for all these items but it is especially important for autonomy, which roughly speaking is the ability to rationally make decisions for oneself and be a “self-legislating will”, i.e. someone who makes decisions on the basis of rules that they impose on themselves. Arguably autonomy involves the capacity for reflection insofar as one cannot automatically or unconsciously self-legislate; to self-legislate in this sense necessarily involves stepping back and reflecting on the type of life one wants to live.
For example, consider the concept of an “advanced directive”, which is a special legal contract that allows people to decide how they want to die. Suppose your friend Alice had never heard of an advanced directive before nor had she ever considered the question of how she wanted to die e.g. whether she would want to live on life support for more than six months. Now if you asked Alice about advanced directives and she responded instantly with a “no” you would be confused. You would say, “How can you answer so quickly? Don’t you need to reflect a little longer on the question?” It would be one thing if she said “Oh, actually I have thought about this before and my answer is still no.” But it would be another thing altogether if she said “I don’t need to think about it – I just went with my gut reaction, and that gut reaction is no.” If she answered in this way you might think she did not understand the moral significance of advanced directives, which demand a certain kind of slowness in deliberation in order to be morally relevant.
Consider another example. You notice your friend Bob has grown really close to his girlfriend, Carol. One day you ask Bob if he wants to marry her and he instantly answers “Yes”. Surprised, you ask, “So you have thought about this before?” and Bob says “No, I’ve never thought about it before until you asked.” Most people would find this strange because marriage is such a significant life decision that it demands slow, deliberative reflection. To not reflect on such weighty issues indicates a failure of moral agency.These two examples illustrate a general principle about the crucial role reflection plays in supporting rational, autonomous choice, namely, that it must have an element of “slowness”. This kind of reflective autonomy is distinct from the autonomy of, say, cats, who are free to choose between sleeping on the mat or sleeping on the bed. The latter kind of autonomy is what we might call sentient autonomy because it’s possessed by almost all Earthly beings that are sentient. Sentient autonomy is important and distinguishes animals from, say, rocks and dust bunnies but it is not the only kind of autonomy relevant to moral agency. If there was a being that possessed reflective autonomy but wasn’t sentient, it seems absurd to deny them moral agency. Reflectively autonomous agents would be able choose to help moral patients regardless of their ability to sensuously feel pleasure or pain. Moreover, their decision procedures would be such that they are of a deliberative nature, grounded in reasons that they are able to explicitly articulate if necessary.
Consider the fictional character Commander Data from Star Trek. Data is an advanced android with a positronic brain that can compute trillions of operations per second. He is thus hyper-intelligent, processing information faster and more accurately than any human. Even if his brain is a computer Data is not merely a computer; he is a moral agent just the same as any human. The only difference is that Data is not a sentient being in the sense that he lacks the bodily consciousness of animals and other fleshy creatures.
Biting the bullet and denying Data moral agency is implausible given that Data was often the wisest and most morally principled of all the crewmembers, not to mention the most valiant in the face of action as evidenced by his many medals of honor. If anyone was capable of reflective autonomy if was Data. It might look from all appearances that he was acting out of just normal sentient autonomy but this is an illusion generated by the sheer speed of his reflective processing. Consider the numerous medals won for bravery and honor in service of Starfleet. All of Data’s valor and bravery were executed not because of any animal instinct or sentient autonomy but because he made a reflective choice. This is evident by the fact that if you asked Data why he performed action X in situation Y he would always be able to explicitly articulate a reason for having done so, even if that reason is “Because I was programmed to do so”. The relevant point however is that his actions betray the flexibility, switching, and autonomy relevant for moral agency as well as the explicitness characteristic of reflective agency.
3. Reflective Moral Patiency
In this section I will defend the second half of premise (2): the capacity for reflection by itself is sufficient for moral patiency. Any entity that can reflect is what I call a reflective patient. The guiding intuition behind experientialism is that welfare flows from the capacity to experience the world, not the capacity to reflect on the world. However, I contend that if there was a being that was insentient but capable of reflection it would be wrong to harm them. Take Data again.I contend that it would be wrong to treat Data poorly by either intentionally destroying him, being negligent to his robotic body, or needlessly destroying his prized belongings. In other words, Data is a moral patient that cannot be treated like just any mere physical object.
There are at least two objections someone might have to Data being a moral patient. First, the experientialist might simply balk at the thought Data cannot feel pain and pleasure. How could his cognitive life be identical to that of a rock or other insentient entities? Surely there is a qualitative or experiential dimension to Data’s existence that distinguishes his existence from that of rocks and dust bunnies. I would respond by saying there is indeed a certain “quality” to Data’s information processing but I’m not convinced we are forced to say such information processing is “experiential” unless that just means “has a quality”, which would trivialize the notion. I can grant the quality of Data’s positronic brain as it reflectively operates is different from the quality of a rock because of its informational complexity without supposing the quality is necessarily due to the information processing being experiential in way an animal’s sensuous pleasure or pain is experiential. In effect, I’m proposing that an entity could have the quality of being a reflective thinker without being a subject of phenomenal experience.
The second objection is that moral patiency plausibly flows from an entity having interests that can either be satisfied or frustrated. Didn’t Data have interests and aspirations like anyone, however “robotic” or “inhuman”? If Data is merely engaging in reflective thought but lacks any interests then the objector might say it’s implausible that his life could be made better or worse and thus would not count as a moral patient. Since we’ve already argued that Data surely is a moral patient then his patiency must be due to a kind of experiential welfare, as per experientialism. The underlying assumption seems to be that unless a cognitive capacity is experienced it cannot be intrinsically valuable and thus cannot be a suitable locus for moral patiency. Call this the Principle of Experience (PE). Kahane & Savulescu also endorse a version of PE writing that “phenomenal consciousness is required if a person is to have a point of view, that is for the satisfaction of some desire to be a beneﬁt for someone” (2009, p. 17). The intuition behind PE is that what makes it permissible to randomly shoot a rock and impermissible to randomly shoot an animal is that rocks lack phenomenal experiences that can be negatively or positively affected.
However, I believe this objection fails to fully grasp the distinction between reflective patiency and sentiential patiency. Data can be a moral patient so long as we are careful to distinguish “bottom-up” interests that stem from animalistic sentience, and “top-down” interests that stem from reflection. It’s debatable whether Data has genuine bottom-up interests but undeniable he has top-down interests due to his capacity for complex, reflective thought. For example, Data might not have a sentential instinct to avoid pain but he can reflectively think “I do not want to be destroyed.” Data could surely sign an advanced directive and his signature would be morally relevant because he can explicitly articulate and reason about his decision. It would be wrong to intentionally destroy or mistreat Data not because he can experience the mistreatment but because it would violate his reflective interest to continue existing. If Data signed an advanced directive it would be wrong to intentionally ignore it for the exact same reason it’d be wrong to intentionally ignore a human’s advanced directive.
Another kind of thought experiment supports the intuition that reflective consciousness is relevant to moral patiency independently of its relation to sentience. Consider the hypothetical scenario where a chimpanzee and a chicken were in a burning building and you could only save one. Other things being equal, it seems overall better to save the chimpanzee because although both the chicken and chimp are sentient arguably the chimp has a greater amount of proto-reflectivity that is intrinsically valuable. Similarly, if the choice was between a chimpanzee and an adult human, it seems overall better to save the human for the same reason: the human is sentient and it is reflective. Furthermore, suppose your mother or father was dying and the doctors said they could save their life only on the condition that they would be insentient but reflective. They would be able to converse intelligibly, write emails, thoughtfully answer questions about their own folk psychology, cook dinner, and otherwise act like perfectly normal people except they couldn’t experience pleasure or pain. Would you accept the offer? It seems absurd not to. The rich, multidimensional intelligence associated with reflection is valuable independently of any contingent relation to sentience. These thought experiments lend credence to the thought that moral status comes in degrees and that reflective moral agents that are also sentient carry what some philosophers call “Full Moral Status” (Jaworska & Tannenbaum, 2013). Moral patients that are sentient only carry less than full moral status because they are not reflective patients.
I’ve argued that experientialism is false because it assumes that all moral patients and all moral agent are necessarily sentient. In contrast I’ve attempted to open up the conceptual space by arguing that the capacity for reflection itself is sufficient for both moral agency and moral patiency.
Bernstein, M. H. 1998. On Moral Considerability: An Essay on Who Morally Matters. New York: Oxford University Press.
Farah, M. J. (2008). Neuroethics and the problem of other minds: implications of neuroscience for the moral status of brain-damaged patients and nonhuman animals. Neuroethics, 1(1), 9-18
Jaworska, Agnieszka and Tannenbaum, Julie, “The Grounds of Moral Status”, The Stanford Encyclopedia of Philosophy (Summer 2013 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/sum2013/entries/grounds-moral-status/>.
Kahane, G., Savulescu, J. (2009). Brain damage and the moral significance of consciousness. Journal of Medicine and Philosophy, 34(1), 6-26.
Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological review, 84(3), 231.
Regan, T. (1986). The case for animal rights. In P. Singer (Ed.) In Defense of Animals (pp. 13-26). New York: Basil Blackwell
Suchy-Dicey, C. (2009). It Takes Two: Ethical Dualism in the Vegetative State.Neuroethics, 2(3), 125-136