“And just as the presence of a verbal report is the best evidence that someone was conscious of something, the absence of verbal report is the best evidence he was not (barring forgetfulness, deception, or mental dysfunction)” (Anxious: p. 150)
Joseph Le Doux is a neuroscientist who has worked on emotions for many years and has made some brilliant advances in the field. While his work within neuroscience has had a huge influence, it has also had a massive influence outside of the discipline in areas like philosophy of mind, psychoanalysis, psychiatry etc. When psychoanalysts and neuroscientists decided to create an interdisciplinary school called neuropsychoanalysis to bring insights from both disciplines together, Le Doux’s work was turned to as providing a nice bridge between both disciplines. For a time Le Doux was even a member of the editorial board of the Journal of Neuropsychoanalysis. Neuropsychoanalysts like Solms, Turnbull and Zellner regularly cite approvingly the work of Le Doux. Furthermore psychoanalyst and philosopher Marcia Cavell in her ‘Becoming a Subject’ uses Le Doux’s research into memory as a way of providing empirical support for psychoanalysis.
Given that Anxiety is a key psychoanalytic area of research a reader with a background in psychoanalysis may expect ‘Anxious’ to deal with some psychoanalytic theories of Anxiety and perhaps provide a bridge between psychoanalysis and neuroscience. While Le Doux does briefly discuss Freud and some existential analysts the book largely stays clear of psychoanalytic theories. In fact there is a strong behavioural bent to the book with heavy emphasis on the role of Pavlovian conditioning in anxiety and scepticism about claims that animals have emotions in the way humans do. When Le Doux discusses consciousness his explanation is similar to the lingua-formal views of Dennett and Jackendoff. From a psychoanalytic point of view it is interesting to note that Le Doux is heavily critical of Panksepp’s claims that all mammals share a similar emotional system. Panksepp is closely involved with Neuropsychoanalysis and a lot of psychoanalysts rely heavily on his work when trying to draw a bridge between the two disciplines. In this respect I think it is very important to evaluate what effects Le Doux’s claims will have on psychoanalysts who use Panksepp’s work as a way of building a bridge between the two disciplines.
Le Doux criticises Darwin’s theory of emotions and the various different emotion theories which we have inherited from those influenced by Darwin. He calls this the standard view and cites Jaap Panksepp’s basic emotional systems: Seeking, Rage, Fear, Care, Panic, Play etc as the modern exemplar of the standard view. Panksepp by studying brain physiology and animal behaviour argued that these basic systems play a role in humans as well as animals. Thus mammals like rats, dogs, and humans share similar basic emotions. Le Doux, on the contrary, argues that the evidence in these cases is consistent with animals engaging in instinctive behaviours but doesn’t necessarily show that non human animals are experiencing the human emotions Panskepp attributes to them. Le Doux notes that the evidence clearly shows threats do release innate behavioural responses and physiological patterns, but the attributing to animals reacting in these ways feelings of fear is an interpretation. He instead champions the view that fear in humans are psychological constructions that strongly depend on our linguistic abilities.
This view is one that Panksepp, and those in Neuropsychoanalysis influenced by him strongly disagree with:
“A plethora of evidence indicates that a non-thinking core-consciousness is generated by extensive sub-neocortical circuits, emanating from brain areas as low as the mid brain. These non-cognitive, non-verbal brain functions are the foundation for our mental existence. They are the seat of the primal, the biological soul-the core self. As already briefly noted, an impressive amount of data indicates that the neo-cortex cannot generate or sustain consciousness on its own. When various sub-cortical structures are damaged, consciousness is simply eliminated. However, when the cortex is damaged early in life, both animals and humans maintain an effective emotional presence in the world. Clearly neocortical skills in processing information entering the brain through sensory portals gradually provide many additional tools of consciousness-the experiences of fine grained vision, acute hearing, to various language skills, and even an appreciation of the nebulous nature of causality- but these neuro-mental treasures are not essential for raw affective consciousness itself. For these reasons we disagree with the feedback and read-out theories that envision affective experience and consciousness in terms of cognition and language.” (Panksepp: ‘Affective Neuroscientific View of Human and Animalian MindBrains’ p. 162)
“When studying animals, we make the assumption that emotional feelings go along with their emotional behaviour because there is no reason to believe (no evidence to the effect) that they do not-and we are therefore comfortable in asserting that by studying the neural correlates of behavioural manifestations of fear, maternal behaviour, addiction and so on, in humans and other animals, we also have a window into studying the subjective affect associated with those states” ( Solms and Zellner: ‘Freudian Affect Theory Today’ p. 135)
It is not a surprise that people like Solms and Zellner agree with Panskepp over Le Doux. One of the primary reasons that Panskepp’s work has been so influential in Neuropsychoanalysis is because in many ways it seems to back up some central claims of Freud. In Solms and Zellner (2012) they note that Freud’s second layer of affect corresponds to Panksepp’s basic emotion systems. Freud, like Panksepp theorised that we have innate behavioural and physiological responses to various aspects of our environment and that these responses are represented in the subject in the form of pleasurable and unpleasurable experiences. They note that there are some differences between Freud and Panksepp, for example, Panksepp’s theory is more explicit and worked out than Freud’s sometimes vague formulations, and it avoids Freud’s untenable appeal to Lamarckism. Nonetheless they argue that if Freud’s theories are suitably modified by Panskepp’s discoveries then psychoanalysis is shown to be in pretty good shape in light of modern neuroscientific discoveries. They note for example that Panksepp’s ‘SEEKING’ system is extremely similar to Freud’s concept of ‘Libidinal drive’. They also link Panksepp’s ‘PANIC’ and ‘CARE’ systems with Bowlby’s emphasis on the importance of Attachment to a care giver for the developing child. Given the fact that they rely heavily on Panksepp’s views as a way of updating and improving Freud the following assessment from Le Doux will be troubling for them:
“The main difference between my view and Panksepp’s is therefore, whether subcortical systems are directly responsible for primitive emotional feelings or instead are responsible for nonconscious factors that are integrated with other information in cortical areas to give rise to conscious feelings. What Panksepp calls cognitive feelings, are I maintain, what feelings are. The subcortical states are, as he also says at times, “truly unconscious” and thus not feelings at all. They are, in my view, nonconscious motivational states.” (Anxious p. 150)
What Le Doux calls nonconscious motivational states is essentially just physiological processes and behaviour it leaves out entirely the actual affects. Panksepp (as well as Solms) defend dual aspect monism a position which emphasises that there is only one substance the physical world, but it can be viewed from two different aspects: from the point of view of first person experience or from third person science. Panksepp notes that while the man on the street will attribute emotional feelings to animals they would be surprised to see that thinkers like Le Doux seem to deny this. He correctly notes that Le Doux’s view equates consciousness with the neo-cortex in particular the language centres of the brain (ibid p. 162), and he thinks that empirical data from neuroscience refutes Le Doux on this matter.
I think that Panksepp’s characterisation of Le Doux is basically correct; but Le Doux may not agree with me on this. Le Doux is careful not to explicitly claim that animals do not have emotions, rather he makes the weaker claim that we have no compelling evidence to think they have other than by an argument from analogy with humans. He claims that while it is possible that animals have emotions we have no clear evidence that they do and furthermore even if they do we have no clear evidence that their supposed emotional feelings are anything like ours. He goes on to offer a variety of considerations which he thinks should give us pause for thought before uncritically attributing emotions to non-human animals. He notes experimental evidence that indicates that people can be shown pictures of threats in a manner that stops them consciously perceiving the threats, and these people report no feelings of fear. Nonetheless their amygdala is lit up by the stimulus and there are physiological responses, such as sweating, increased heart rate, elevated pupil size etc. He takes this to be evidence that humans can have automatic physiological responses to threats which are not accompanied by conscious feelings (Anxious p. ix). He notes we have no reason to think that non-human animal’s innate behavioural responses are conscious as opposed to being variants of the nonconscious physiological response to threats we see in humans in experimental settings. He further argues that if we say that animals have consciousness like humans we will run into a problem of deciding where to draw a line; he asks rhetorically where do we draw a line are cockroaches, worms, and crayfish capable of experiencing anxiety or are they just unconsciously implementing mechanisms to avoid danger? He argues that the simplest solution is to assume that we have inherited the ability to detect and respond danger from animals and not feelings like fear. He notes that if we agree that the ability to detect and respond to danger does not require consciousness then we will not have to debate whether and which creatures from worms to dogs have conscious feelings of anxiety. He concludes as follows:
“The evolutionary function of this ancient capacity is not to generate emotions like fear or anxiety, but simply to help ensure that the organism’s life continues beyond the present…Survival circuits do not exist to make emotions (feelings). They instead manage interactions with the environment as part of the daily quest to survive” (ibid pp 43-44)
I think it is interesting to note that while Le Doux begins by stating that he is not denying that animals have feelings like fear, happiness etc his arguments end up reaching the stronger conclusion that evolution has fitted us out with an evolutionary ancient ability to detect danger as opposed to feeling fear. He later goes on to argue that feelings of fear are connected to our conceptual capacities:
“Ultimately, feelings like fear require that we somehow have the concept of fear, based on words and their extended meanings, in our minds…Animals obviously cannot label and interpret survival circuit activity in ways made possible by human language. They may experience something, but it is incorrect, in my opinion, to assume that their experience is the same as, or even similar to, what a human often experiences when a defensive survival circuit is active in his or her brain.” (ibid p.46)
Note in the above passage we have a bit of slippage. He equates feelings of fear with having the conceptual abilities derived from the public meaning of words; yet he later makes the weaker claim that we should be sceptical of claims that non-human animals have feelings similar to ours. It seems that the logical conclusion from his first claim should be that since non-human animals do not have public shared meanings they do not experience fear at all. So clearly Le Doux’s claim that he is not asserting that non-human animals do not have emotional feelings is at odds with the general tenor of his arguments which do indeed make this strong claim. This tension persists throughout anxious on the one hand he seems to want to deny that animals have conscious experiences, then he implies that he is only making a weaker claim that we cannot know if they have feelings. His disavowals aside I think his own arguments commit him to the view that non-human animals do not have emotional feelings. Hence his serious disagreement with Panksepp’s theory of a basic emotional systems shared by all mammals. Another curious feature of Anxious is the casual way Le Doux attributes propositional attitudes to non-human animals which is at odds with his careful treatment of attributing consciousness to non-human animals. The quote below is instructive:
“Specifically, when animals engage with their environments in daily life, they rely not only on innate responses and conditioned reactions, but also on goals, values, and decisions, each of which involves complex cognitive processes but which do not necessarily require conscious awareness of these processes…But scientific practice, at least in my view, cautions us, in the absence of rigorous and compelling evidence, to avoid the attribution of conscious feelings to animals on the basis of our intuition, no matter how strong, that animals should have such feelings” (ibid pp47-48)
Note Le Doux’s stringency in deciding whether to attribute conscious states to non-human animals but his casual manner of attributing goals, values, and decisions to non-human animals. Debates about whether animals have concepts, intentionality etc are massive at the moment with both scientists and philosophers debating the issue vigorously. Philosophers and scientists like Chater, Hayes, Dummett, and Davidson have all presented arguments that we are not justified in attributing concepts to animals while others such as Stich and Fodor have argued that they do have concepts. As we saw above Le Doux argues that that to have a feeling of fear requires having a concept of fear which he thinks is somehow derived from public language which animals do not have. Yet if he argues that an animal can have goals, make decisions, and have values this seems to imply that it has some concepts independent of public meanings. Why a non-human animal can have goals and not fears is not specified. It is possible that Le Doux just means that non-human animals have as-if intentionality which we attribute to animals because it helps predict and control their behaviour but they have no real intentional states outside of the attributions of language users. It is hard to know precisely what he means but on a simplest possible reading it seems that he is inadvertently treating intentionality attributions to animals with less care than he treats attributions of consciousness to animals.
Le Doux is heavily critical of Panksepp’s theory of universal basic innate emotions shared by all mammals. Given the emphasis neuropsychoanalysts place on the supposed fact that Panksepp has provided justification of key Freudian concepts; Le Doux’s criticisms of Panksepp will have consequences for neuropsychoanalysis. Obviously I cannot here go into the entire Freudian theory and evaluate every aspect of his theory which Panksepp seems to support, and how Le Doux’s criticisms will effect these concepts. Instead I will focus on a psychoanalytic treatment of anxiety which is built upon Panksepp’s work and I will discuss how if Le Doux is correct his criticisms will bear on the project. Finally I will evaluate whether I think Le Doux’s criticisms of Panksepp stand up to critical scrutiny.
In 2012 Yoram Yovell gave a talk to the neuropsychoanalytic foundation called ‘Anxiety: One or Two? He began his talk by discussing attachment theory and some of the ways it is studied. He described a typical experiment as follows: At 12 months a child is observed playing for 20 minutes, the mother leaves the room and the child’s reaction is recorded and analysed. The mother and child are together in a room with many toys to interact with. There is a stranger in the room (usually a woman), the stranger doesn’t interact with the child. The mother and child play together for a while then without warning the mother gets up and leaves the room. Usually the child begins to cry when this happens. Then the stranger come and tries to console the child by playing with him. The mother returns. Then the stranger leaves. Then the mother leaves this time leaving the child all alone. The child fends for himself.
Yoram argues that test is close to a biological variable in that the behaviour of the child in this experiment is highly predictive of the child’s behaviour later in life. Ainsworth mentioned 4 main types of attachments (Yoram only discusses 3)
- Secure Attachment: When the mother leaves the child cries and protests. But when the stranger comes the child plays with the stranger and is happy. Upon the mother returning the child runs and hugs the mother and plays with her.
- Insecure Resistance: After the mother leaves the room the child becomes very upset and is not consoled by the stranger. After the mother returns the child hugs the mother but is even more upset and cries hysterically.
- Avoidance: The mother leaves the room the infant goes on playing doesn’t look upset. When engaged by the stranger the baby plays with her. But later on when the mother returns the child ignores the mother and continues playing with the stranger.
Yoram notes that all three patterns of attachments are considered normal. The child’s behaviour in these experiments are highly predictive of future relationships. Having outlined the experimental paradigm he goes on to discuss Freud. Freud argued in these situations the child feels physical pain when the mother leaves. He asks what is the source of this physical pain which distinguishes it from other types of anxiety? He argues the threat of the loss of an object triggers the anxiety and he distinguishes this type of anxiety from what he called pure anxiety which occurs when the stimulus that provokes it is not a beloved object. Freud’s view of anxiety was that it can originate from several different sources. It manifests itself in two different ways painful anxiety and non painful anxiety these are caused in different ways. Separation anxiety and fear of loss of love is what causes painful anxiety.
Yoram noted that Melanie Klein expanded on Freud’s views on anxiety. Her discussion of anxiety, like Freud’s broke anxiety into two key types:
- Paranoid Anxiety: Fear of Annihilation.
- Depressive Anxiety: Fear of loss of a beloved object or lover.
Yoram argues that when we look at the history of psychoanalysis we see that there are two main positions on anxiety. Many psychoanalysis believe that anxiety manifests itself in many ways but is one thing e.g. Bolwby, Rank, Winnicott, Kohut. However Klein and her followers Meltzer, Segal, etc believed that there were two different types of Anxiety.
Yoram thinks that this dispute can be clarified by looking at pain, anxiety, and object loss from a neuropsychoanalytic point of view. He discusses Japp Panksepp’s four distinct emotional command systems which we share with our fellow mammals. The Fear System and The Panic System are related to Anxiety according to Yoram. He points out that from a point of view of neuroscience there are similarities between human and animal emotional systems.
- The Fear System is an ancient system that is found in reptiles. It appears to be an all purpose alarm system that warns of existential threats snakes, predators etc. Fight or flight system. With an emphasis on flight. It is triggered by the lateral nuclei of the amygdala. The central and lateral nuclei of the amygdala are involved in the fear system and the medial nuclei, and basil nuclei are involved in the rage system. The activation of these systems can be inhibited by benzodiazepine drugs e.g. ativan.
- The Panic System: It is concerned with Anxiety, Pain and Loss. This system is newer than the Fear System. It mediates attachment relations for organisms. Yoram notes that the Panic System maps on very beautifully to Bolwby’s Attachment System. It mediates the fear the child develops when separated from his mother. It is connected to the thalamus, the ACC and. Its actions can be inhibited by the SSRI antidepressants Prozac, as well as by narcotics.
He notes he has worked on narcotics as anti depressants and that one interesting finding is that narcotics are potent anti-depressants and they are the strongest known pain medications. He argues that the contemporary evidence indicates a common overlap between the registering physical pain and social pain (this is similar to Panksepp’s Panic System). This is very relevant to Freud’s linking of Pain to Separation Anxiety.
He goes on to argue that the Anterior Cingulate Cortex is important in separation anxiety and in physical pain. Cingulotomies are sometimes done on people with conditions that are so painful that even narcotics do not work in relieving pain; they are also sometimes done as a last resort to help people with intractable depression. In Panksepp’s terminology people who got cingulotomies underwent a legion in the Panic System which helped them deal with their symptoms. A consequence of cingulotomies are reduced concern for social connection, less caution in social interaction. It makes you (a) less anxious of social rejection (b) less dependent on social bonds (c) it does not touch fear of fire, fear of heights etc.
Yoram further claimed that you can have a lesion in the Fear System without having a lesion in your Panic system, and we know this because of Damasio’s patient S. She had her amygdala destroyed because of a genetic disorder. She lost the ability to recognise fear and anger on other people’s faces. The same lack of recognition of Fear and anger in all modalities sound, stories etc. But S could recognise all other emotions in people. She lost the ability to tell suspicious people from trustworthy people. She wanted to form a connection with anyone who was interested in talking to her. She inappropriately hugged and touched people who she had little reason to trust. As a result she was taken advantage of a lot over this. She had the ability feel loss and pain at being mistreated but still maintained a need to form connections to these people. She lost the fear of all other daily activities snakes, fires, etc.
Yoram notes that based on descriptions by people like Freud and Klein we seem to have two different types of Anxiety. (1) Depressive Anxiety (2) Paranoid Anxiety. Based on lesion studies we can see that one of these systems can be damaged without damaging the other. The two systems respond to medications differently; the fear system responds to Benzodiazepine while the Panic system responds to endorphins, narcotics SSRI’s so people like Klein and Freud were correct in hypothesising that there were two types of anxiety.
In his talk Yoram obviously didn’t go into all of the technical details of the techniques used to discover the neurophysiological details of the so called FEAR SYSTEM and PANIC SYSTEM. A substantial amount data was drawn from studies of non-human animals and these studies assume that non-human animals have emotional systems very similar to ours. This assumption is of course the very one that Le Doux is calling into question. The obvious point to make is that since Yoram above was relying on studies of brain damaged humans then his hypothesis is not at odds with Le Doux’s view. It is theoretically possible that the brain damaged systems of the people Yoram discussed are systems are not the innate emotional systems that Panksepp discussed but are systems that have developed partially through humans conceptual and cultural systems. In this case the evidence he presents would not be at odds with Le Doux’s views. At this moment in time all the evidence we have indicates that all mammals share the same general neural architecture and that damage to either the FEAR or the PANIC system results in patterns of behaviour consistent with Yoram’s data. The only difference is that in the case of humans we have linguistic reports to go along with the behavioural data. Ultimately while it is theoretically possible to make data presented by Yoram, Solms, Panksepp consistent with Le Doux’s theoretical framework there seems little need to do so, as the evidence is not on Le Doux’s side.
Le Doux offers a variety of different arguments against the attribution of feelings like anxiety or fear to other mammals. He notes phenomena like people having physiological and behavioural responses consistent with fear in certain experimental settings but who report having no subjective feeling of fear. Dennett who holds a similar view makes the same argument about people with blind-sight. But these arguments only tell us that it is theoretically possible that animals are have physiological and behavioural consistent with fear but no subjective feeling of fear; but the argument does not give us any reason to believe that this sceptical scenario obtains. It is true that it is a logical possibility that such a scenario obtains, just like it is logical possibility that solipsism is true, or scepticism about other people having minds is justified. But in all these cases we need more than a mere logical possibility to justify the claim that animals do not have feelings we need some empirical evidence to convince us that this is the case.
Le Doux does offer some arguments to support his scepticism. He argues, for example, that if we claim that mammals are conscious then we end up with a problem of saying where the dividing line is between which creatures are consciousness and which ones are not. He asks us rhetorically do earthworms have anxiety? He has a point we really do not have a good handle on where the dividing line between conscious and unconscious processes are. Obviously we have a good handle on what Le Doux calls creature consciousness e.g. being awake or alert. The difficulty arises when we consider mental state consciousness, e.g. to be aware that one is experiencing something ( ibid p. 150). He thinks that creature consciousness involves being awake and alert, responsive to sensory stimuli, execute complex behaviours, solve problems and learn from past experience. Mental state consciousness has all of the features of creature consciousness but has further features such as (1) Awareness that a stimulus is present (2) Awareness that one’s self exists, (3) Awareness that it is oneself that is sensing, behaving, and solving problems ( Anxious p. 147). Le Doux though notes that when we try to decide which creatures have mental state consciousness we are bombarded with possibilities. Some possibilities being:
- Only humans have consciousness.
- Only primates have consciousness.
- Only Mammals have consciousness.
- Some invertebrates have consciousness.
- Consciousness is biological information all living creatures including plants and unicellular creatures have consciousness.
- Anything that integrates information in a particular way (cell phone, internet) has consciousness. ( Anxious: 148)
Those familiar with the literature will be aware that there are a variety of competing positions from outright panpsychism to eliminativism about consciousness. The truth is at the moment there is little to decide between positions like Deacon’s which argues that autogens may have proto-sentience and rivals like Integrated Information Theory which speculate that sentience may be possible for even smaller units. At the moment we can only really say for certain that humans and creatures with a similar physiology to our own (other mammals) have consciousness. Whether creatures further from us on the evolutionary scale have consciousness is something that is more difficult to tell. But an epistemological boundary which makes it difficult to draw the line about which creatures are conscious should not be used as an excuse to draw an arbitrary boundary which states that only language using creatures have consciousness. At the moment we have no reason to doubt that mammals are consciousness and therefore I don’t think Le Doux’s book really poses any challenge to Panksepp’s work.
 For a good introduction to this topic see Kristin Andrews (2015) ‘Animal Minds’.
 For evidence that fish have the ability to feel pain see Victoria Braithwaithe’s ‘Do Fish Feel Pain?’ (2010)