Michael Tye on Do fish have feelings?

Michael Tye is of the view that the problem as to whether fish have feelings or not, is best addressed epistemically than metaphysically. At his lecture at King’s College London yesterday entitled ‘Do fish have feelings?’, he compiled and appraised the available empirical research and made his important conclusions. I would have preferred him to merely assert that science has demonstrated fear, anxiety and pain responses in many species of fish – both reactive behaviours and adapting responsive behaviours – and then to have moved swiftly on by exploring philosophical questions such as, ‘do these fear, anxiety, and pain responses equate to feeling?’ ‘does it matter that fish feel pain?’ or ‘how painful is a fish’s pain?’ However, he did express the view at the start that, particularly in the USA, many remain unmoved to the primary question and need battering over the head with the evidence. (Alternatively, you can read this article as a pdf).

Michael Tye argues that fish experience anxiety fear and pain

Michael Tye is correct in saying that fish have feelings.


One of Michael Tye’s interesting views which was expressed briefly at the close, was that in contrast to fish, insects do not feel. Clearly, he believes that there is some kind of transition from organisms that have no feeling to those that do. This reminded me of the problem I have identified in Daniel Dennett’s ‘Intentional Stance’ c.f. ‘Intentionality: Dennett’s 1 vital error is Searle’s 1 critical omission’. So I asked Michael Tye a question to the effect,

is there a greyscale of degrees-of-feeling for things like pain that correlates with degrees of organism complexity (or of neural complexity), or is he of the view that there was a sudden evolutionary jump – a step-change – that granted feeling to certain organisms and not to others?

His answer entailed referring to one particular species of jellyfish. Apparently, when a single pain ‘cell’ of this species of jellyfish is stimulated, it leads to the sympathetic stimulus of all its pain cells. There is no small pain and no degree of pain, but rather just an all-encompassing on or off pain.

On the evolution of the precursor to ‘feeling pain’

Survival pressures dictate that replicating organisms will tend to evolve mechanisms that mitigate harmful environmental effects. The term “harmful” can be equated to any ‘bad’ environmental effects that restrict survival potentials. Thus in the simplest of replicating organisms there is a potentiality gradient: inevitably, this gradient is realised through the evolution of increasingly sophisticated mechanisms that are responsive to the qualitative relevance of environmental stimuli because to do so is of survival benefit. Consequently, over the evolutionary timeline, increasingly sophisticated mechanisms evolve that enhance the conflation of environmental experience and improve the kinds of mechanisms that institute reactive behaviours. However, capabilities such as these need not be neural. Rather, they can come about through any bio-chemical representation or apparatus that leads to responsive actions to mitigate harmful effects – which goes some way to explaining the evolution of neurone and neural mechanism precursors. Even very simple organisms – by which I mean such things as unicellular creatures – will have mechanisms that cause physicochemical responses to ‘bad’ environmental experience; they will show aversion to noxious stimuli (an aversion that has evolved because of its tendency to mitigate effects that are harmful). Thereby, one can say of the physiological construct of such creatures, that it possesses a representation of the qualitative relevance of environmental conditions via its complex biochemistry. This representation amounts to a ‘knowledge’ about the environment; by virtue of its tendency to cause reaction in a manner that, in some way, accurately reflects the nature of environmental conditions. This is the case even if the expressed behaviour is of a merely innately acquired and reactive kind.

On the evolution of phenomenal feeling and mentality

But would such limited reactive responses be indicative of the presence of the phenomenon of feeling? Can such complex physicochemical representations be interpreted as constituting mental phenomena? In answer to these questions, despite the incredible sophistication of these biochemical mechanisms, I would say no:

Whilst there is undoubtedly purpose and intent behind the mechanisms of unicellular creatures for instance, one is hard-pressed to describe these mechanisms as generating mental events with feelings attached, or to describe the responses they elicit as “behavioural”. One is inclined to think therefore, that the ascription of mentality tends to be reserved, out of default and therefore rather bluntly, for organisms that possess a neural cluster and network “of-any-description“. Michael Tye was quick to point out that looking at a fish brain and seeing no neocortex is, similarly, a blunt and unqualified method for concluding that fish must have no feeling.

So what then constitutes mentality and when can we say of a neural cluster and network, that it creates phenomenal feeling rather than merely innately acquired reactive responses to stimuli? Well, there is a clue in the question: “merely reactive responses”:

Innately acquired mechanisms lead to unalterable reactive consequences. These kinds of ‘consequences’ – particularly those that are implemented with no neural involvement but rather by only biochemical means – indicate fixed representations about the qualitative relevance of environmental conditions that have evolved over generations. There is no process that can be called mental, that grants individuated conditional and considered responses to realtime experience.

In contrast, as soon as an organism is capable of assessing and prioritising the qualitative relevancy of experience as and when it happens – that is, as soon as it adjusts reactive responses to cater for realtime variations in types of experience – it is able to mitigate innate reactionary consequences and respond with ‘some degree of behavioural sensitivity’ to particular realtime scenarios. This capability is reflected in an organism’s tendency to act in a manner that is most appropriate for survival in the given situation i.e. a tendency to do ‘what it feels right to do’ in response to noxious experiences at that particular time following a considered response to that particular scenario. Hierarchical Construct Theory indicates that an organism that does this, is feeling something about its changing environment and responding as it feels it should – “it” being an individual rather than an innate automaton. Such organisms demonstrate responsive rather than merely reactive behaviours. They are observed to learn from experience and to adjust responses to the nature of changing experiences. Thus they possess a responsive individuality and can frame their biochemical knowledge within the bounds of a developing evolving understanding regarding the qualitative relevancy of experience. The second by second evolution of understanding (by this definition) is a realtime response to the qualitative evaluation of experience and, importantly, is an individuated response rather than one that is dictated by innate evolutionary adaptation. This realtime responsiveness is a behaviourally adaptive phenomenon that leads to the experience of an ever changing understanding, specifically, regarding the qualitative relevance of experiences – be they good or bad. In other words, this realtime changing interpretation of the qualitative relevance of experience is the phenomenon of feeling experiential relevancy, and hence can be regarded as constituting a mentality.ñ

Afterthoughts about Michael Tye on concepts

A central thesis of ‘Seven Puzzles of Thought’ (Tye, 2012), is that concepts are constituents of thoughts. It is difficult to argue that processes that generate an evolving understanding of the qualitative relevance of experience (as above), do not constitute ‘thinking’. Such processes require at minimum sophisticated neural mechanisms and a substantial neural network that can conflate sensory experience, then assess, and evaluate it, and finally prioritise responses. However, I contest, such processes need not generate conceptual interpretations of the qualitative relevance of experience.

Hierarchical Construct Theory indicates that non-human animals do not introspect about their phenomenal state, which is why they do not attempt to communicate their conclusions to this introspective analysis. They experience the phenomenal state – which entails thought – in the absence of conceptualising about it. Thus, they can discriminate phenomenal characteristics and form associations between phenomenal characteristics and environmental events, and they express the experience phenomenon through gestures and utterances that communicate the qualitative realtime impact of experience, but such things do not entail conceptual constructs.

Concepts arise from a higher degree of relations – a higher degree of representation regarding the interpreted value of environmental influences. Concepts are a construct of principled relations. Thus to simply learn, is not to understand the principles behind learning: whilst non-human animals learn, they do not try to learn. Learning is merely a by-product that arises both from the assimilation of phenomenal feelings, and an immediate association with their environmental causes. Nor do non-human animals attempt to improve their learning through due process. Similarly, to experience the phenomenon of feeling the effects of experience, is not to understand the principles of phenomenal consciousness nor to conceive of one’s self as the principled object to which the feeling is felt. Thus, with just phenomenal feeling, one need not conceive of the self as its recipient. Rather, one just experiences the feeling and responds as one feels one should; as one thinks best. Humans are the only animal with an evolving understanding of the principles regarding the value of phenomenal experiences and thereby are the only animal to feel compelled to speak of this unique realisation regarding the nature of their reality. This realisation induces the creative exploration of the principle relation between the phenomenal qualitative aspect of experiencing reality and the nature of its cause. Such developing realisations are themselves part of a complex conceptualised, and therefore, self-aware and self-identifying construct.

Feel free to respond… thanks.

Leave a Reply

To leave a comment, please prove you are a person by answering this CAPTCHA