This is a very interesting article on word embedding in current vector-thought research:

Word Senses polysemy

The article quotes a 1957 chapter by Firth: ‘Firth’s hypothesis that a word’s sense is captured by the distribution of other words around it.’

Here is Firth’s original text:

Firth linguistic theory synopsis

The so-called Firth’s hypothesis sounds very Saussurean to me. Saussure is usually credited with the idea that language is a system of differences, where each atom’s meaning  is given by its place in the structure. Hence my idea that thought vectors are a reactivation of structuralism for new purposes. It seems that post-WWII structuralism was itself influenced by cybernetics, but I’d like to investigate Saussure’s influences.

 

 

 

Neural networks have the rather uncanny knack for turning meaning into numbers. Data flows from the input to the output, getting pushed through a series of transformations which process the data into increasingly abstruse vectors of representations. These numbers, the activations of the network, carry useful information from one layer of the network to the next, and are believed to represent the data at different layers of abstraction. But the vectors themselves have thus far defied interpretation.

Read the article here: Decoding the Thought Vector

The purpose of this post belongs to the process of determination of a specific crealectic space of knowledge and examination, if any. I’d like to proceed in concentric circles, starting with the smallest possible unit. This would avoid dispersion and allow, perhaps, a clearer grasp and approach regarding the field of crealectics: what it might or might not cover. The question will of course not be solved in a single post, especially in the exploratory phase that this blog is meant to represent. 

I shall start with a heuristic hypothesis regarding the smallest unit of study in crealectics. Let’s venture that it could be the mark. If crealectics is about the different logoi that can be articulated out of the Creal material (previously defined as infinite disparity or multiplicity), one might be tempted to claim: the smallest unit of a logos is indeed a mark. I mention the mark rather than the sign, because a sign seems to be an evolved category of mark, one that supposes a referent, the something that it stands for. One could perhaps imagine a mark that is not a sign, that does not signify (and hence is not a signifier either). But perhaps this is mere word-play.

The Oxford dictionary proposes, as the primary definition of mark: a small area on a surface having a different colour from its surroundings. The keyword here is difference – colour is one perceptive possibility. Now the first criteria of distinction of a mark seems to be precisely that it is remarked, noticed This seems to position crealectics as a discourse on perception.

It might be that there is no such thing as a mark in itself, an independent mark. It might be, as the phenomenologists would have it, that a mark is always a mark for a remarker. And, as the structuralists would have it, that a mark is always part of a series of marks. How?

Consider for example a white tiny grain on the table. I notice it – we could say that it is the sign that something exists there in front of my eyes, even if I don’t know what it is (a grain of sugar?). But before it signifies an existence, a thing, it is a mark, indeed a white-coloured mark on a green table. 

Now before we dive into the cognitive neuroscience of colour perception in bees, it might be worth remaining at the level of phenomenological analysis a few seconds more. A mark is a distinction. It is part of the cognitive feature of a mark that it is remarked. To notice a mark, I need to distinguish something from something else. But do I make/co-create the distinction or is the distinction fully real, there, independent of my perception? Science would say that, even if we now know that colours are partly a co-creation of the brain, the distinction is not fully abstract, it is mostly a property of the objects being different in structure. 

One could say, again in phenomenological or structuralist fashion, that the mark is nothing in itself. What makes it a mark is the difference with other marks. What is a difference? It is a relation of inequality or non-identity, real of perceived. What one should consider as the unit of crealectics would not be the mark as an entity or thing, not even the fact that a consciousness or Husserlian cogito remarks a mark, but that a difference is instantiated. There would be no distinction without a difference, and there would be no mark without a variation (notice here en passant that if a mark is a difference, a sign could be called a differing, alluding here to Derrida’s somewhat structuralist suggestion of a difference that differs, a différance).

Now, as announced, I’d like to pause my speculation by looking at a paper entitled ‘Small Brains, Smart Minds: Vision, Perception and ‘Cognition’ in honeybees’ (Mandyam Srinivasan & Shaowu Zhang (2003), IETE Journal of Research, 49:2-3, 127-134).

Why? Because the paper claims, after a series of experiments, that ‘bees can learn to navigate through labyrinths, to form complex associations and to acquire abstract concepts such as “sameness” and “difference”.’ In another paper, more or less the same team of scientists explain (M. Giurfa, S. W. Zhang. A Jenett. R Menzel & M. V. Srinivasan, The concepts of”sameness” and “difference” in an insect, Nature, vol 410, pp 930-933. 2001):

An important cognitive capacity is the ability to learn relationships between stimuli. In vertebrates, the capacity to acquire sameness and difference concepts has been studied using two experimental procedures, the matching task and the oddity task. A variation of the former is the ‘delayed matching-to-sample’ task, in which an animal is presented with a ‘sample’ and subsequently with two or more secondary stimuli, one of which is identical to the sample. The animal is required to respond to the stimulus just encountered. The ‘delayed non-matching-to-sample’ task is a varia- tion of the oddity task. The procedure is similar to the matching-to- sample task except that the animal is required to respond to the stimulus that is different from the sample. In both cases, broadly construed sameness and difference concepts are shown only if the animal exhibits positive transfer to a completely new set of stimuli, which it had not experienced during training.

The authors conclude that bees possess the concepts of difference of sameness.

Bees live according to a series of robust protocols which are a given articulation of the Creal, a living system, or, to use Wittgenstein’s famous phrase, a form of life. Having ordered the Creal in a certain way does not mean that this ordering was intentional or conscious in a human-like way. But still, the slow elaboration of a living system or bio-logos seems to imply precisely the constant definition of differences and samenesses. The concepts of difference and sameness are constitutive of what it is to become a living being with a regulated Umwelt. What the above papers says is this: non-human agents of a differential perceptive system have the capacity to distinguish variations, and that capacity is dynamic (can be trained with new stimuli). But distinguishing variations is part of the genesic phase of the system. It should therefore not surprise us to find this cognitive capacity operating once the system is up and running.

The question is: are the bees presented with radically new information, or simply stimuli that correspond to familiar perception capacities, even if organised in a different way? The scientists have used colours and odours, which are not outside of the realm of what a bee can usually perceive (even if some colours can be new). One could argue that the scientists have only played with signs or marks that are familiar to bees, if in an unfamiliar way. The question is: could a bee perceive a mark outside of its usual realm of perception? Or could humans perceive something that is not part of their perceptive apparatus (not a sound, not a colour, not a word, not a symbol, etc.)? If we only perceive what our species has learned to perceive, in other words, if cognition is perception-dependent, then crealectics would be, like biosemiotics (and not very differently), a branch of cognitive biology. 

But what if biology were a subset of cosmology and metaphysics. If the Creal is posited discursively as the source of crealectics, then two directions of speculation are open: what if we (humans and bees) were and are differentiated in a cosmic system that encompasses all particular systems and forms of being, like marks in a wider programme? Perhaps we can distinguish marks because we are nothing but marks ourselves. In this case, crealectics would be a form of neo-structuralism.

Or: what if the axiomatic positing of difference as difference were itself different from the distinction of difference? In that case, saying that bees perceive differences does not mean they have the concept of difference as difference. And positing the concept of difference or infinite probability would be — rather than the mark — the origin of crealectics, as I have tentatively proposed in this paper. In such case, crealectics would not be about biology primarily, but about politics and ethics.

 

 

Science of Logic part I

Philosophy or science is after-thought. Thought is primary, not reflexive. Thought is life.

“philosophy should understand that its content is no other than actuality, that core of truth which, originally produced and producing itself within the precincts of the mental life, has become the world, the inward and outward world, of consciousness. At first we become aware of these contents in what we call Experience.”

Thought is actuality, which is deeper than appearance.
Spirit is the cause of the world
The only absolute or infinite objects are God, Spirit, Freedom
“thought in its very nature is dialectical, and that, as understanding, it must fall into contradiction—the negative of itself—will form one of the main lessons of logic.”

“thought is one out of many activities or faculties of the mind, co-ordinate with such others as sensation, perception, imagination, desire, volition, and the like. The product of this activity, the form or character peculiar to thought, is the UNIVERSAL, or, in general, the abstract. Thought, regarded as an activity, may be accordingly described as the active universal, and, since the deed, its product, is the universal once more, may be called a self-actualizing universal.”

“The difference between conception and thought is of special importance: because philosophy may be said to do nothing but transform conceptions into thoughts—though it works the further transformation of a mere thought into a notion.”
“Particular ends can be attained only in the attainment of what absolutely is and exists in its own right.”

On the Concept of Creal: The Politico-Ethical Horizon of a Creative Absolute

 

Luis de Miranda

Published in The Dark Precursor: Deleuze and Artistic Research, edited by Paulo de Assis and Paolo Giudici, Leuven: Leuven University Press, 2017

 

Process philosophies tend to emphasise the value of continuous creation as the

core of their discourse. For Bergson, Whitehead, Deleuze, and others the real is

ultimately a creative becoming. Critics have argued that there is an irreducible

element of (almost religious) belief in this re-evaluation of immanent creation.

While I don’t think belief is necessarily a sign of philosophical and existential

weakness, in this paper I will examine the possibility for the concept of universal

creation to be a political and ethical axiom, the result of a global social

contract rather than of a new spirituality. I argue here that a coherent way to

fight against potentially totalitarian absolutes is to replace them with a virtual

absolute that cannot territorialise without deterritorialising at the same time:

the Creal principle.

Read the full paper here:

Miranda_Luis_de-On The Concept of Creal