All The Things

The desire to define “everything” can be traced back thousands of years in our known history. Aristotle, in “Categories”, provides us with one of the first recorded attempts of uniquely categorising all “things” – both animate and inanimate, even directly contradicting the views of his teacher, Plato, while doing so. In parallel to this, more than three thousand miles to the east, in the Vaisheshika and Nyaya schools of the Hindu philosophy, the “Padārthas” were created – first verbally and then in writing – in order to serve a similar purpose. These geographically – and temporally – separate efforts, aiming to achieve the same general goal, show us that there is an innate need to conceptually organise everything that constitutes our perception of reality, in a categorically distinct manner.

Many western philosophers have since attempted to redefine and expand what Aristotle set in motion, with the intention of further developing the way that we specify and distinguish things. However, as the cognitive linguist George Lakoff (1987:pp.6–7) points out, the classical theory that acted as the “seed” of this movement was solely based on “a priori speculation” – not on empirical study, yet it has been traditionally taught as an “unquestionable, definitional truth”, instead of a hypothesis. Before we can begin to challenge our understanding of categories, however, we first need to challenge “understanding” itself.

The linguist Benjamin Lee Whorf (1952:p.21) states that “the why of understanding may remain for a long time mysterious; but the how or logic of understanding […] is discoverable”. He goes on to say that the reason we decide on specific ways to “cut up” and organise things and events, is because “we are parties to an agreement to do so, not because nature itself is segmented in exactly that way for all to see”. This is further highlighted by the fact that languages do not solely differ in a syntactical or grammatical manner, but also in how they “break down nature to secure the elements to put in [their] sentences”, suggesting that speakers of each language perceive reality in different ways – a hypothesis which has been researched and supported further since (see Giang, 2018).

The mathematician Frank Ramsey spent a significant part of his paper entitled “Universals” arguing against Bertrand Russell’s theories on particulars and universals. Particulars are unique entities, such as, e.g., a specific person, whereas universals are shared properties, such as characteristics or qualities. Ramsey (1990:p.13) proclaimed that even though Russell recognises philosophers are commonly misled by the “subject-predicate” construction of the English language, even Russell is in fact misled by language itself, since “the whole theory of particulars and universals is due to mistaking for a fundamental characteristic of reality what is merely a characteristic of language”.

The philosopher Susanne Langer (1933:pp.179–182) offers a different perspective to this concern. She first adds Russell’s proposition that “there is a further element, the relationship or structure” which, when put together with the various other elements – like the subjects and the objects, forms our “logical language”. She then proposes that even a single language can produce a variety of different systems of understanding, but all systems “may be reduced to a highly abstract level” consisting of “arbitrary formulations”, thus “no structure is absolute, no relation peculiar to the material in hand, no analysis of fact the only true one”. Perception of reality hence lies in the eye of the beholder.

It is an interesting realisation to consider that we formed logical language by verbalising our understanding of the elements, and we are now trying to go back to the “source” by attempting to “reverse engineer” the logical language that we previously formed. In this search for the origins of meaning, we start building models in an attempt to formulate our knowledge, and rely on cognitive psychology theories to help us recognise why we perceive things in the way that we do.

When we start designing our cognitive models, we may inevitably find ourselves realising that, as computer scientist Allen Newell (1994:p.54) stated, “[k]nowledge abstracts from representations, yet knowledge must [still] be represented in some fashion in order to be used, told, or thought”. But how can we tell in which ways to correctly organise and represent this knowledge? The psychologist John Robert Anderson (1990:p.95) finds that the creation of a category can be predicted via linguistic labels, which offer indications that it exists, as well as via the similarity of features and functions of objects, which enable its definition by incorporating these shared characteristics. When we decide that an item falls under a specific category based on its attributes, it allows us to identify additional attributes of this item that we had not realised before, which derive from the characteristics that we had pre-assigned to the selected category. However, if we find that one of these extra attributes is not a correct match for this specific item, this shows us that we must define a new category in order to contain it.

The cognitive scientist Zenon Pylyshyn (1985:pp.38–40) maintains that our interpretation of a model and its characteristics could merely be considered as “attributions” made by whomever is creating or viewing it. This brings into question whether the identified semantics are the “original” ones, or are instead derived from the interpretation itself. The latter would be problematic since, as Pylyshyn argues, such derived semantics “cannot literally cause a system to behave the way it does; only the material form […] is causally efficacious”. Fortunately, there is no need to counterargue by invoking the “observer effect” – viz. how we can change a phenomenon simply by observing it, since Pylyshyn goes on to say that the human brain, just like a computer, derives these semantics based on the “physical properties” of the model, which infers that these properties are “psychologically real” and thus actually hold meaningful “semantic content” that can potentially help us pinpoint any misunderstanding.

The scholar Alfred Korzybski (1994:p.xvii) supported this concern, declaring that every map is also “a map of the map-maker: her/his assumptions, skills, world-view, etc.”, once again highlighting that we must take representations “with a grain of salt”. Korzybski (ibid) is most famous for his statement that “the map is not the territory”, which aims to express the common fallacy of our human tendency to confuse a visualisation – e.g. a map or diagram – with the actual thing that it represents. However, he further adds to this point that “no map represents all of its presumed territory [since] maps are self-reflexive, i.e., we can map our maps indefinitely“. This is indeed a “slippery slope” that needs to be carefully managed by the designers of any visualisations that aim to effectively inform or educate their audience.

Competent teachers can present the same material in many different ways, until they manage to get their intended message across. They are able to do so because they thoroughly understand what they are trying to communicate, and can consciously – or even unconsciously – “see in their mind’s eye” how each concept relates to other concepts, thus becoming capable of describing it via alternative methods and pathways. We all become producers and consumers of visualisations on a daily basis, every time we design or draw something, or when we view someone else’s creations. But how can we be certain that the intended message – if it exists – is indeed getting across?

Education, in its current state, uses a wide variety of sources in order for teachers to help students learn. From written works and graphical representations, to audio narrations and video animations, there is an endless plethora of learning resources created by teachers for students to use while studying specific topics. However, one can often find that these teaching material contain certain embedded – yet silent – assumptions. For example, reading these paragraphs alone indirectly presumes that you comprehend written English to a specific degree, are acquainted with Harvard-style citations and references, and even know how to operate a digital device that is capable of viewing electronic documents. One who has not mastered at least the aforementioned skills – and the many other skills “concealed” within these, would potentially be unable to even begin to consider reading this essay. If their intention were to also understand it, then they would need an entirely new collection of expertise, relevant to the various theories of philosophy, psychology, and education, in addition to all of the above.

It is becoming increasingly more obvious that what education needs is not just new learning theories and teaching material, but also ways to identify the parts that make the whole, in order to be able to realise what we are truly presented with. By breaking concepts down to their very discrete components and looking at how they are interlinked, we can potentially identify how they are also interconnected with other concepts – to certain degrees, hopefully enabling us to pinpoint the “hidden prerequisites” that may be keeping students from grasping the concept at hand. Once this is in place, perhaps students will be able to take ownership of their learning and overcome any perceived boundaries, on their way towards grasping the desired lesson.

We thus need to separate the content from the structure, in order to be able to express the same message in different ways, according to the needs of our students. By having our content clearly discretised and planned out, we could become able to produce it and share it with others in a collaborative manner, e.g. via online communities that aim to exchange “Learning Objects” (see Koper et al., 2004) in order to be reused and further expanded. This can open up new horizons, potentially enabling us to achieve more effective knowledge dissemination, with the aim to minimise miscommunication and maximise understanding.

Frank Herbert (1982:p.311) wrote in Dune: “Deep in the human unconscious is a pervasive need for a logical universe that makes sense. But the real universe is always one step beyond logic”. Even if we never manage to create a single “ultimate” collection of interconnected objects, representing the “absolute” truth, due to the inherent limitations of our present form of existence, we can still aspire to venture outside established conjectures, and verge towards unexplored horizons, with the hope to discover, acquire and propagate newfound understanding, for and to all.



Anderson, J.R. (1990) The Adaptive Character of Thought. Lawrence Erlbaum Associates.

Giang, V. (2018) How language shapes our perception of reality. Fast Company. [Online]. Available from:

Herbert, F. (1982) The Great Dune Trilogy: Dune, Dune Messiah, Children Of Dune. Book Club Associates.

Koper, R., Pannekeet, K., Hendriks, M. & Hummel, H. (2004) Building communities for the exchange of learning objects: theoretical foundations and requirements. Alt-J. [Online] 12 (1), 21–35. Available from: doi:10.1080/0968776042000211502.

Korzybski, A. (1994) Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics. 5th edition. New York, Institute of General Semantics.

Lakoff, G. (1987) Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. The University of Chicago Press.

Langer, S.K. (1933) Facts: The Logical Perspectives of the World. The Journal of Philosophy. [Online] 30 (7), 178. Available from: doi:10.2307/2015751.

Newell, A. (1994) Unified Theories of Cognition. [Online]. Harvard University Press. Available from: doi:10.1126/science.3291119.

Pylyshyn, Z.W. (1985) Computation and cognition – Toward a foundation for cognitive science. The MIT Press.

Ramsey, F.P. (1990) Philosophical Papers: F. P. Ramsey. D. H. Mellor (ed.). Cambridge University Press.

Whorf, B.L. (1952) Collected Papers on Metalinguistics. [Online]. Department of State, Foreign Service Institute. Available from: