First written: Dec. 2014; last update: Dec. 2014
This paper introduces the theory of Panmnemism as a variation of panpsychism. Rather than making claims about spirit or sentience existing in all things, panmnemism references the ability of inanimate objects to store experiential information as memories. Memory, or mneme, is to be understood as the capacity to store information concerning the experiential qualities of existence. Experiential information can be stored as memory. Memories can be retrieved and communicated to investigators. From these claims, it will be concluded that non-human animals, plants and inanimate objects possess a form of memory, as well as forms of semi-linguistic communication of the information within those memories. Viewing the world in this way allows all entities to contribute to the development of knowledge. Such contributions lend purpose to all things.
It has too often been the case that discussions involving references to panpsychism stray into fears of repellant mysticism or the tendency to encompass aspects of spirit or animation. There is a simpler, more specifically targeted argument which can be adapted from panpsychic theory, one which will likely avoid those metaphysical traps. Instead of attempting to support the existence of cognition or soul in non-humans, the focus ought to be on the capacity to collect memories and construct narratives. The qualia of ‘being in the world’ is attached to the existence of individual objects. The ability to store subjective information and pass it on at a later time is a fundamental quality of consciousness. This is only one aspect of mental function, but it’s one that can be attributed to inanimate objects with little contention. Panpsychism, at its core, has always claimed that some aspects of mind must be present in all things. This idea can be traced back at least as far as the Epicurean notion that the will cannot emerge out of nothing, and thus must be present in all forms of matter. Understanding consciousness as having existed prior to the appearance of human life is a necessary element of Buddhist and Hindu cosmologies. Numerous scientists, from Thomas Edison1 to David Bohm, have argued in favor of panpsychic ideas in order to explain the behavior of forces and particles. Leibniz and Spinoza both built philosophies dependent on the belief that mind and spirit are composed from smaller blocks of mental attributes, which are present in all things. Precisely which elements of mind are present in matter is rarely specified.
Philosophical claims of emergence refute this idea of consciousness covering of a wide spectrum of proto-cognitive abilities, suggesting that consciousness appeared in a sudden evolutionary leap. To understand adaptation as a process is to understand that whatever abilities are present in the human animal today must have begun, in more primitive forms, in our distant past. The intricate operations of consciousness could not have emerged from out of nowhere. Consciousness must have evolved from simpler instances of conscious activity, or proto-consciousness.
Any arguments stemming from mind/body debate partisanship are irrelevant to the discussion of consciousness as categorized by panpsychic theory. If dualism were true, the secondary immaterial substance of mind could easily be assigned to all physical things, not just human brains. Since this mental substance is non-physical, and thus immeasurable, it could be active around any form of matter. Conversely, if the identity theory holds true, panpsychism would still have ample space in which to operate. If consciousness is reducible to molecules and matter, then all things constructed from molecules and matter would possess the potential for similar functions. This is not to suggest that inanimate objects are capable of experiencing emotion or actualization, though not all humans are capable of those particular cognitions either. Since emotions, for one example, are not exhibited by all thinking humans, their presence can’t be a necessary condition of consciousness.
Just as the term consciousness has been continuously redefined, the word panpsychism has seen itself applied to all sorts of theories and meta-theories. In order to avoid any pluralistic confusion, I will be using a term which more specifically handles the aspects of mind being addressed here. Panmnemism is the position that all things have memory. This clarification immediately brackets out all implications of animism, vitalism and substance dualism. Memory is not alive. Memory is not magical. Memory is not a secondary substance. Memory is the act of storing information with a capability for future retrieval.
Panmnemism is a variation of panpsychic theory, but dodges the problem of proving self-awareness; the trait often relied upon as the only sufficient condition for sentience. This is not to say that panmnemism is merely a watered down or compromised version of panpsychism. The discussion of all things carrying the potential for remembering and transmitting experiential information has utility. In fact, gathering information and storing it for later retrieval might have as much practical value as other aspects of conscious ability. Even more beneficially, this view of material nature helps to remove the explanatory gap of emergent experience.2
Experience is an attribute of existence. All physical entities are items which exist in the material world. To be within this spatio-temporal reality is to be effected by aspects of the surrounding environment. Where something is located, how it grows or degrades, changes in the atmosphere around it are all types of direct events acting upon singular perspectives. Such events ought to be called experiential. Each object is unique in its location, so each object develops experiences which are specific to its position, size and relation to its surroundings. These individualized traits can be referred to as subjective experiences. If these events are not experiences or proto-experiences, one is left concluding that there was no such thing as experience prior to the development of neural systems. Pattern, color and location would have had to suddenly emerge from nothingness, after billions of years of non-existence, once a sentient being got around to noticing them. Self-awareness is not a prerequisite of existence. Whether or not an object has the ability to reflect upon the quality of its experiences is irrelevant to the fact that physical entities are shaped and impacted by their place in the world.
There is a need to clarify what separates panmnemism from previous attempts at narrowing the scope of panpsychism. David Ray Griffin and Gregg Rosenberg endorse a reduced panpsychism of their own, known as panexperientialism. This theory, largely indebted to Whitehead, holds that all objects and entities gather individuated experiences.3 While this is certainly an important element of the philosophy being discussed here, their theory is quick to dismiss the presence of any sort of mentality in those entities. Experience is treated as a physical certainty, but is disassociated from the activities of minds.4 This harsh delineation reduces the attributes of inanimate objects to little more than what physicalist philosophies grant them in the first place.
Panmnemism, on the other hand, allows all objects the opportunity to participate in the collection of experiential information and the growth of a universal knowledge base. While the act of introducing newly fabricated terms into the philosophical lexicon tends to raise suspicious brows, this specific set of conditions cannot be confused with the more inclusive (panpsychism – all things are conscious) or the overly exclusive (panexperientialism – all things are experiential). Mneme, from the ancient Greek, refers not only to memory, but the ability to retain environmental information. Mnemic Theory, now little more than a relic, proposed the idea that memories were inscribed in the protoplasm of living cells as engrams.5 This is not what panmnemism is proposing, though the underlying concepts are similar. Panmnemism also encompasses the possibility for inanimate forms of expression. By exhibiting the power to share experiential information, the objects of our world participate in the assembly of knowledge, rather than keeping their experiences locked up privately.
As a physical and retrievable element of being, experiential data have great importance to the accumulation of knowledge. Not all experiential data are cognitive experiential data. Value stems from the quality and quantity of information, not from the carrier entity’s awareness of that information. It is through the collection of information that observations and conclusions concerning nature, ethical progression and evolution can be ascertained. If knowledge has a value for its ability to drive efficient practice and ethical improvement, then what is best for the greater good would be that which successfully preserves a record without causing harm or otherwise disrupting the developmental process. If all objects can carry extractable information then all objects can be said to assist in the growth of knowledge. A system for assigning value to an object, based on the potential contributions of that object to the expanding catalog of experiential data, can be laid out with a basic set of criteria.
Before setting up a framework for evaluation, the role of externally stored information in our progress and evolution ought to be explored. The knowledge base accumulated from our world is consistently utilized as the primary measurement of progress. Whether we are discussing technology, health and wellness, philosophy or even global politics, the distance gained from previous epochs is measured by a growth in understanding. Collecting information from the natural world is paramount for this evolution. Becoming more rational, or even more enlightened, is a reliant on intellectual progress, directly or indirectly derived from observations of matter. The ability to uncover data is expanding. The ability to store data is expanding. Advancement is linked to progress in the endeavor to gather and assess facts and figures from the widest variety of sources.
Without going into too much quantitative detail, a foundation can be established to appraise this value based on the following factors: experiential memory storage potential, ease of information extraction and a pragmatic assessment of the quality of the data received. First off is the potential for memory storage. When looking at computer storage, encoded information in cells or the fossil record, we are dealing with very large sets of extractable information. The storage capacity of an object or entity is a valuable space for information. It is real estate for future knowledge. Like any other storage facility, the value is based on the capacity, not the contents. It is the potential for knowledge which must be credited rather than actualized data. If we were to look at a parallel on the human level, we wouldn’t want to exclude the potential contributions of the majority of first person perspectives, simply due to poverty or socio-political marginalization.
The next variable under consideration relates to the ease of extraction. If data collectors are not aware of the experiential knowledge inside of an object, or they’re unable to access the object’s memory, the utility of that knowledge sinks toward the bottom of this value scale. Systems for extracting packets of information, such as radiometric dating or tree ring analysis, are well established research tools. These are examples of relatively easy information extractions. Several new transfer systems are being experimented with. Molecular memory, insect communication and environmental toxicology are areas undergoing current exploration. The ease of extraction will only be advanced by the creation of a reliable process and a common language for sharing any information retrieved. It might be beneficial to associate these extractions with the term recollection, rather than memory. Memory, for better or for worse, belongs to the vocabulary of human psychology. Recollection, the act of collecting again, is a fitting descriptor for these examples. The object collects experiential information and that same information can be re-collected at a later date.
Lastly, we come to the weighing of the epistemological value of the information itself. Information storage which can be kept static for long periods of time, without tampering or alteration, has greater value than consensus-based historical recording. As human history progresses, the narratives shift, the detail sometimes fades and politics directly or indirectly censor the neutrality of the information. The same factors apply to the extraction and interpretation of individual panmnemic data points, but not to the raw content of those data points prior to human interference. The way in which people interpret century old climate statistics varies. Any inferred meaning associated with that data can be swayed by scientific bias, political motivations or ordinary human error. But the data keep their value, as long as they are returned to a state of static information, which can be examined and employed again and again through future analysis.
A star sends valuable information, about the physical and environmental conditions in its galaxy, across the universe. The experience of a particular celestial body has its greatest value as potential knowledge. When humans get involved in hypothesizing and speculating, the value decreases. A sensor, collecting the raw data, skips the middle-person’s conjecture and keeps the potential for the transportation of valuable knowledge at its highest. It has become a fairly common practice, or at least a convenient tool, to think of elementary particles as information.6 This is the function of our interaction with a majority of substances, we engage with them as little more than robust information sources.
Experiential information derived from non-humans has the asset of being untainted by false or implanted memories. People not only have difficulty recollecting past events with objective detail, they also carry in their brains all sorts of perceived impossibilities, lies, errors in education, dreams and imagination. A story my sister told me about an accident I had when I was two years old seems as real to me as any other recollection from my early childhood. If that tale was invented, and its truth value was withheld from me, the memory would contain the same quality inside my mind. The false data are passed off as genuine every time I retell the story as if it had actually taken place. Rocks and molecules appear to be incapable of lying. This grants a higher value, or at least a greater reliability, to whatever stories they pass on to their investigators.
These might be relatively primitive levels of experience, but they can be found in all entities. Memory and experience do not need to mimic human memory and human experience in order to have a value. If the information derived from an object serves a purpose, then the object has performed a useful function. Experience evolves toward complexity, from tiny independent units to intricate and collaborative systems which provide immense bodies of viable data to whatever is able to extract that information. We should not necessitate the presence of complex systems with collaborative functions in order to refer to the collection and dissemination of data as recollection. That would lead to problems of where to draw the boundaries and whether a sentient being with damaged neural functions still possesses memories. Again, this is not proposing an equivalency between memory and consciousness, but suggesting recollection as an intermediary in the evolution of consciousness, or as a type of proto-experience.
In order to explain the appearance of cognition in nature we need to place current levels of conscious ability within an evolutionary context. Mind did not emerge from nothingness. The suggestion of mental emergence is counterproductive to our growing understanding of natural selection. The picture becomes more subtle, explicable and closer to completion if we adhere to Teilhard de Chardin’s position that biological evolution must be a continuation of pre-biological evolution.7 Incorporating a few other panpsychic theories will help to solidify the additive process I am describing. German biologist Bernard Rensch wrote about the proto-phenomenal properties of inanimate matter.8 He explained the necessity of attributing such characteristics to atoms and molecules, or unnecessarily suffering through the alternative, trying to integrate consciousness into some unknown intermediary level. If we combine this concept with key elements from the panpsychic writings of Gustav Fechner, specifically his ideas about mind resulting from a systemic layering of simple material functions,9 we begin to see a more inclusive evolutionary model. This additive view can be further quantified with the application of a physics of information. The smallest components of the natural world are not necessarily quarks or strings. All sub-atomic particles can be discussed as being made up of measureable experiential data sets. Wheeler’s theory of ‘It From Bit’ explains that, “every particle, every field of force, even the space-time continuum itself – derives its function, its meaning from apparatus-elicited answers, to yes or no questions, binary choices, bits”.10 From this we can see how even something storing its history of, for example, receiving light versus not receiving light, would be sufficient to represent that data as a binary number set. This is usable information which takes on meaning once it has been transmitted or observed.
Memory, as discussed within the frame of panmnemism, is to be understood as the ability to store information concerning the experiential qualities of existence. The element of choice or selection is irrelevant. Just as people can’t decide which details to remember and which ones to forget, other material things hold on to some aspects of their experience in the world while some pieces go unrecorded. Higher order comprehension still carries a greater value, though, that preference does not reduce the importance of remembering and sharing memories as a form of proto-consciousness. Being able to recall and restate facts and figures will get a student passing grades in many courses. While educators would prefer to create a system which requires a deeper understanding of the subject matter, it can’t be said that rote repetition has no utility. Demonstrating defensive measures to offspring, based on past predation, has an important function in animal survival. That’s regardless of whether or not the animal has any knowledge of what they’re teaching or why they’re teaching it. People carry memories of sensations that they weren’t able to comprehend when their minds first received them. We can remember things that still don’t make sense to us, even after years of cognitive reflection on the primary experiences. Human memory is fallible and largely unidirectional. If I can’t recite a phone number backwards as well as forwards, or recall it without picturing the keypad, do I really have a firm memory of the number? Examples of rote information dispersal descend way beyond higher order animals.
Slime mold demonstrates detailed patterns of path recognition and spatial memory. It will avoid searching for food in places it has already checked and sort out the shortest paths to and from food sources. This is an organism with no neural function.11 Meteorites carry information about their stellar origins. Hair can be examined to reveal a detailed history of forensic toxicology. Transmission of stored experiences continues to develop. New material areas are being explored for the potential historical information lying dormant inside.
Wooden musical instruments for example, have resonant top plates which vibrate in specific frequencies. Certain nodal patterns, correlating to the commonly used pitches of the performer, are repeatedly flexed. This repetition stiffens some areas of the structure while loosening others. As the wood slowly ages and dries over time, these Chladni patterns are set as sympathetic nodes in the object.12 If these past preferences are able to be extracted, then it would be wrong to say that the plane of spruce on the top of a violin responds as though it remembers being played in down-tuned, just intonation. If the wood expresses familiarity with those frequencies, it does remember, and has shown evidence of its past experiences by resonating in that mode.
There is a peculiar tendency to rank mental faculties hierarchically. As thinking human beings, we often place our own complex brain functions right up at the top of that pyramid. This practice is likely a matter of protecting our species from any threat to our cerebral supremacy. Admittedly, panpsychism, panexperientialism and panmnemism present a comparatively neutral, passive view of conscious action, but one which Descartes yearned to reach, one which Buddhists believe to be highest. Why is cognition considered to be of a higher order than qualia? Why is self-awareness believed to be a more evolved trait than experience? Why is creativity valued over recollection?
Without a consensus based definition for what consciousness is, philosophers are left discussing the various elements which constitute conscious thought. Multiple traits and abilities might figure into the assembly: self-awareness, rational thought, recollection, language, empathy, moral judgment and creativity are all functions of a conscious mind. No single trait is sufficient to represent the presence of sentience. Consciousness can best be understood as a convergence or overlapping of a number of these spheres. Essential to this Venn diagram are the spheres of communication and memory.
Memory plays an essential role for identification by providing a constant for the continuity of consciousness. The first-person narrative of any individual is dependent upon reliably referencing identical memory points within that individual’s history. What is self-awareness without memory? Because I can recall the same specific childhood fishing trip I was able to picture five years ago, ten years ago and twenty years ago, I can be sure that the events which have taken place over those twenty years fall within the same life narrative. This elevates the forming and saving of memories to a high position among the necessary cognitive abilities of sentient beings. This is the formative basis for conscious reflection as well as the requirement for maintaining a continuity of self.
Language is commonly listed as a necessary condition of sentience. In order for these examples to be considered recollection, the objects need to be able to linguistically express their memories. Objections on the grounds of the language criterion for consciousness can be dealt with by making a slight shift in how language is understood. Language need not be verbal, written or even directly targeted to be categorized as communication. Language can’t be limited to tongue and text. A broader appreciation of linguistic comprehension ought to include communication through non-traditional formats over longer periods of time.
We can personify, even anthropomorphize buildings. People make frequent references to mood swings and attitudes of smart phones and computers. Still, it doesn’t seem right to state that those products possess a language. In order to reconcile this, we must expand our use of the term language from being an immediate interactive system, to also including systems of experiential storage, which can be opened in order to share certain unique experiences. A person with no external communicative ability would still have the internal voice needed to catalog their experiences and emotions. If we think of this one key function of consciousness, the storage and utilization of external stimuli, then conscious ability appears in all sorts of things, not just human minds.
When a tree collects data and stores them in the formations of its rings, this is akin to creating an internal memory bank. All of that environmental information can be extracted later on, providing the recipients with a detailed record of that tree’s experiences. The fact that objects record data in formats which can be harvested and understood shouldn’t be underappreciated here. The object owns its individual information history and there is no clear reason why that history should be transferable to other types of beings. Examining the history of weather, records of trauma, droughts or regional fire activity embedded in tree rings gives the reader of those signs a picture of key events in the life of that organism. This is the tree’s way of communicating its story. Whether or not there’s an intended audience is beside the point; this transfer of experiential information is a type of language.
A more contestable example, but a bit more fun at the same time, might be a consideration of what the French call terroir. This refers to the long and proud tradition of associating flavor profiles with their geographical origins. Burgundy tastes different from Bordeaux, though they’re both red wines. When someone tastes a regional cheese, it is the topography, the soil, the varieties of grazing grass and the weather they are sensing, more than the influence of the farmer or production methods.13 This is why a person with a finely tuned palette can discern between varieties of wine and cheese. They are, in a sense, tasting the story of the grape, the narrative of the milk.
Now, I make no claims about my own ability to discern subtle flavor data and I could stare at rings on a tree stump for days without gaining any useable historical information. This does not mean that the content isn’t present. The narrative exists and is readily accessible. The language must be understood by both parties. Likewise, I’d have no understanding of a page of text presented to me in phonetic Wintu. However, the possible contribution to global knowledge on that page shouldn’t be eliminated due to a lack of linguistic abilities on the part of the receiver. It would be wrong to say that the Wintu language is not a form of communication just because only a handful of people possess a working knowledge of it. There are far more people with a working knowledge of the language of dendrochronology. At least since John Von Neumann’s computational work of the late 1940’s and early 1950’s, the notion of an inanimate object remembering and communicating information has been rationally presented. It shouldn’t be difficult to extend his reasoning beyond machine computation, or to associate such examples of communication with other forms of language.
An experience does not need to have greater meaning derived from it in order to become an important piece of data. Contexts and applications can be associated with those direct experiences later on. This process works in agreement with John Searle’s claim that things don’t have intrinsic intentionality.14 Literary criticism and psychoanalysis are obvious examples of attributing meaning to pieces of information that don’t express a clear perspective in and of themselves. These are impure, or at least imperfect, attempts at a useful exchange of ideas. The philosophical problems associated with language, ever present in post-Wittgensteinian thought, come with the interference of human interpretation. Raw mathematical data actually bring us closer to analytic clarity; they bring us closer to Wittgenstein’s position that description ought to take the place of explanation.15 After all, humans, in the perspective of greater timescales, are little more than carriers of genetic data, data of which we’re not consciously aware.
Humans communicate private experiences of stress or pleasure without a specified linguistic ability to express the quality or quantity of those feelings. I can only compare my pain to other pains I’ve felt, having no calibration for my sensations on a chart which plots the pains of others. These are limitations. The communicative abilities of objects are quite different. The data transferred from inanimate objects and simple organisms are built to specific standards. There is regularity. This makes the transfer of knowledge more reliable. Humans use a language which allows for lies, withholdings and numerous opportunities for miscommunication. In the information storage systems of objects, ‘A’ always equals ‘A’. This type of non-linguistic communication is a more reliable source for the foundations of knowledge. This type of memory is perhaps more analogous to what we label as ‘photographic memory’ in humans. When people can recollect dialogue verbatim or exact dates and names from insignificant events, we not only find that skill impressive, but usually see it as a sign of intelligence. In order to cope with the problems of the future, reliable information taken from the perspectives of all entities needs to be gathered, collated and appreciated.
Developing a better comprehension of the evolution of consciousness, from simple memory retention to complex reasoning systems, will lead to the proper placement of human minds on the widest conceivable spectrum of consciousness. This will allow a future machine or alien intelligence to exceed human conscious abilities without threatening the relevance of human contributions to a universal knowledge base. By imagining mental ability as a vast scale, rather than belonging to a single species, the problems associated with artificial cognition become a matter of evaluating the experiential information in combination with the utility of higher order reflection. There is room for different types of consciousness contributing different strengths of cognition. The likelihood of a scenario with some similarity to Ray Kurzweil’s vision of ‘the singularity’ seems inevitable. The boundaries between organic and inorganic mental states are becoming blurry. As people implant more and more artificial or externally formed devices into their bodies, computers are becoming more organic and beginning to process information more creatively. Even if a fully realized new form of cognitive consciousness is not on the horizon, a very real ‘Ship of Theseus’ type of problem is imminent.
There isn’t a hard line between human consciousness and other forms of consciousness, but there is a range of different aptitudes. Some ‘minds’ are better at preserving detail. Other ‘minds’ have a proclivity for sharing information. Some ‘minds’ excel because of the vast amount of data they can safely store. Human minds are best at organizing and drawing conclusions from data sets.
When the human animal no longer walks the earth, all that will remain of our experiences will be artifacts and a wide variety of extractable information. This historical record will be encoded in thousands of different languages, numeric charts, geologic formations, fossil remains, mathematical proofs, symbolic logic and all sorts of examinable evidence awaiting investigation. People don’t just want to believe that their experiences have a value, we are certain that our perceptions, and the information we gather, can be useful in numerous applications. This knowledge is not only of utility to the lives of future human beings, but valuable as experiential data with the potential of being transferred to any capable investigator. If we choose to ignore the importance of similar examples, where experiential information is being derived from objects today, then there is no reason to place any importance on the experiential information of human beings either. People derive meaning and a sense of purpose from the ability to share knowledge with the rest of the world, and potentially, future civilizations. There is value in carrying information, a value which can be extended to all material data carriers. Rather than dealing with the scientific, philosophical and spiritual traps of talking about consciousness, we can expand our allowance of certain types of conscious abilities to include all contributors. All things have experiences. Experiential information can be stored as memories. Memories can be retrieved and communicated to investigators. From these claims, it can be concluded that non-human animals, plants and inanimate objects possess a form of memory, as well as forms of semi-linguistic communication of those memories.
- Edison, T. “Intelligent Atoms”. As reprinted in The Monist. “Panpsychism and Panbiotism”. Vol. III, 1893. (back)
- The type-1 explanatory gap, sometimes referred to as “Levine’s Explanatory Gap”. Levine, J. “Materialism and Qualia: The Explanatory Gap”. Pacific Philosophical Quarterly. 64, 1983. (back)
- Rosenberg, G. A Place for Consciousness, Probing the Deep Structure of the Natural World. Oxford: Oxford University Press, 2005. (back)
- Rosenberg, G. A Place for Consciousness, Probing the Deep Structure of the Natural World. Oxford: Oxford University Press, 2005. (back)
- Bousfield, W.R. The Basis of Memory. London: Keegan, Paul, Trench, Trubner & Co, 1928. (back)
- Chalmers, D. “Facing up to the Problem of Consciousness”. Journal of Consciousness Studies, 2 (3) 200-219, 1995. (back)
- Teilhard de Chardin, P. Toward the Future. Trans. R. Hague. New York: Harcourt, 1973. (back)
- Rensch, B. Evolution Above the Species Level. New York: Columbia University Press, 1960. (back)
- Hawkins, S. “William James, Gustav Fechner and Early Psychophysics”, Frontiers in Physiology 2:68, 2011. (back)
- Wheeler, J. A. (W. Zurek, ed.) Information, physics, quantum: The search for links, Complexity, Entropy, and the Physics of Information. Redwood City: Addison-Wesley, 1990. (back)
- Reid, Latty, Dussutour, Beekman. “Slime Mold Uses an Externalized Spatial Memory to Navigate Complex Environments”. Proceedings of the National Academy of Sciences. Vol. 109 no 43, 2012. (back)
- Jansson, E.V., Moral, J.A. and Niewczyk, J. J. “Experiments with free violin plates". CAS Journal Vol. 1 no 4, 1988. (back)
- Trubek, A. The Taste of Place: A Cultural Journey into Terroir. Berkeley: University of California Press, 2008. (back)
- Searle, J. Mind, Language and Society: Philosophy in the Real World. Basic Books, 1999. (back)
- Wittgenstein, L. Philosophical Investigations (3rd edition). Trans. G.E.M. Anscombe. Pearson, 1973. (back)