NECESSITY OF GENERALISATIONS
Contents:
A lifeform, to survive, must do two things: grasp reality – in some form – and act accordingly to sustain itself. To elaborate, anything that lives faces an alternative: to exist or not. In the face of this alternative, it must grasp the facts to grasp – in some form – the results of its actions, so it can grasp – in some form – what course of action is most efficacious in sustaining its own life.
Consciousness begins with the most basic form of extrospection: sense-perception. Consciousness is consciousness of something, and thus, the basis of awareness is the irreducible interactions (irreducible in terms of mental units) between the senses and the world beyond the consciousness. Sensations, however, are knowledge only of the moment. Even perceptions, which are sensations integrated in a certain order (automatically), only inform us about the here and now, i.e. that which is present right in front of us.
The limits of such knowledge are clear. The law of identity indicates that observed parts of reality can exist – i.e. that can retain their identity – in other contexts beyond the one we are in. The law of causality indicates that the nature of observed entities can tell us about not only how they are acting but also how they will act beyond the present. Thus, there is more to know than what we perceive, but we can come to know this based on what we sense and perceive, i.e. based on the facts that are in front of us. Given that one’s own life is the standard of value, and given that one’s survival is not a given and must be pursued in reality using one’s grasp of reality, it follows that expanding one’s grasp of reality is the primary way of expanding one’s ability to live, from surviving to thriving.
Furthermore, since we can extend the laws of identity and causality to interactions between entities, and since there is but one reality, everything in reality is interconnected, i.e. integrated to the rest of reality. More specifically, everything exists in a context (i.e. a set of facts conditioning the nature of its existence), and thus, is a part of an integrated system of things. Consequently, every piece of knowledge also exists in a context (a cognitive context involving mental units such as other pieces of knowledge, concepts, assumptions, etc.), and thus, is a part of an integrated system of knowledge. This points to the necessity of integrating knowledge, i.e. grasping the relationships between our pieces of knowledge within the relevant context (relevance being decided by our purposes and our other knowledge). The greater the context we can deal with (great in terms of depth and breadth), the more we can know about what to expect from reality, and thus, the more we can know how to apply our knowledge with efficacy.
Knowledge at the sensory-perceptual level is automatic, and associations between pieces of knowledge at this level can only be concrete-bound, i.e. based on retained concrete observations. Hence, at this level, every piece of knowledge is related to only those pieces of knowledge that are together a part of a broader, concrete observation. At this level, we cannot grasp the broader context that conditions our knowledge, and thus, we also cannot grasp what our knowledge implies for the future.
Hence, we see the value of unit-perception and concept-formation. Unit-perception is the awareness of things as members of a group of similar things, i.e. the awareness of units. Unit-perception enables us to treat similar things similarly without needing to observe everything from scratch. For example, if we observe that stones are dense enough to be thrown with force while leaves are not, we do not test this for every stone or leaf that we come across. Here, each stone and each leaf is a unit, i.e. a member of a group of similar things. In the same way, we can differentiate between different classes of units by differentiating the characteristics shared within each class of units from those shared within the other class. This reduces the number of mental units required to compare and contrast entities, enabling us to relate things in reality more effectively.
Concept-formation depends on unit-perception, but takes it a step further. In a concept, each class of units is retained by not just the similarities between the units but their distinguishing similarities, i.e. the shared characteristics of the units that serve to differentiate them from every other class of units in a certain context. The optimal method of concept-formation ensures that these distinguishing similarities are the essential characteristics of the units, i.e. the basic set of characteristics that imply every other similarity between the units and serve to distinguish these units from other classes of units in a certain context. This ensures unit-economy, i.e. the retention of the maximum cognitive context with the minimum cognitive elements. Concept-formation is not complete until the units are integrated by a single cognitive element (ex. a word, symbol, phrase, etc.) that serves as the label for the cognitive context required to grasp a certain class of units in a certain context. This last process is crucial, as it lets us retain the cognitive context and the hierarchy of abstractions generated during integration as a single, perceptual cognitive element, thus increasing unit-economy even further.
Concepts enable us to see universals in particulars, i.e. to identify the universal aspects (i.e. characteristics and relationships) that unite several – potentially unlimited – particular instances (both known and unknown). Hence, using concepts, we can relate potentially unlimited classes of things with each other, i.e. we can generalise, thus expanding the range of our cognition far beyond what we can sense and perceive. Of course, since only particulars exist as such in reality (universals being derived from particulars), we can only observe particulars. Therefore, to generalise means to induce, i.e. to find – or try to find – universal truths derived from particular observations. Now, note that while a concept identifies an unlimited class of things, it does not by itself give us any more knowledge about this class or its relationship to other classes beyond what is required to identify its members. In other words, a concept is only a way to classify based on the facts, but not by itself a way to extrapolate from the facts. A concept implies any knowledge that has been or can be gained about the class, but it may not be associated with it in our knowledge to begin with. For example, “human” is a concept, whereas the fact that the human heart pumps blood is a generalisation that, once learned, is associated by us with the concept of “human”.
SIDE NOTE 1: Generalisations serve as the basis for further concept-formation, and thus, more advanced concepts, as generalisations provide more material to abstract and integrate.
SIDE NOTE 2: Concepts are a way to classify, but this is a secondary property. They are primarily integrations of abstractions (drawn ultimately from sensory-perceptual data).
Hence, we see that induction (the process of reaching generalisations) is the fundamental method of grasping, i.e. attempting to know reality beyond sense-perception and the abstractions derived from it. Even deduction requires generalisations to begin with, since deduction is the method of showing a conclusion to be self-evident (i.e. logically consistent) within a certain set of generalisations and observations. Note that without generalisations, deduction would be impossible and we would have only observation. Hence, we see that generalisations are central to the use of reason, i.e. the faculty of identifying and integrating sensory-perceptual material.
As rational animals (i.e. animals with the capacity to reason), we have the capacity to conceptualise, generalise, and thus, grasp a context of potentially unlimited scope (with respect to size and time). Moreover, as rational animals, we need to conceptualise and generalise to reason even at a basic level above range-of-the-moment survival. To illustrate this, given below are generalisations at different levels of cognitive advancement:
In order to pursue the value for life to the fullest, i.e. with maximum efficacy in the long-range, it is necessary to recognise the central role of concepts, generalisations and reason as a whole in leading our lives. Of course, just as the pursuit of life is not unconditional, the pursuit of knowledge in a life-oriented value system cannot be unconditional either; we need to choose where to focus our time, energy and resources wisely (this is a topic for another time). Conceptualisation and generalisation enable us to grasp the nature of things in relation to each other, thus helping us understand underlying factors and causal relationships between attributes, entities and events, which enables more (1) effective action in the present based on knowledge of what is more important, i.e. essential to a desired result, and (2) projection of present observations, actions and events to results in a larger time frame (i.e. further into the future).
Furthermore, conceptualisation and generalisation enable us to grasp a wider cognitive context with fewer cognitive elements, which enables (1) more knowledge about how an action or event would be conditioned by other factors, and (2) insights into other areas of knowledge and their interrelation with what we already know. Thus, to live long-range and to grasp as much of the world as we can, conceptualisation and generalisation are indispensable.
Integration of knowledge – of which generalisation is a major part – is essential in expanding a sentient being’s range of efficacy in both thought and action, since it gives a means to deal with a wider range of contexts. To deal with a particular context still requires the particular effort of applying broader concepts, generalisations and principles in the context, but such an ability to apply knowledge is made possible in a wider context by integration alone. Considering a wider context is essential if we do not want to be limited (potentially to a dangerous degree) by a particular context that – being more concrete-bound – can be subject to change. Furthermore, trying to learn for each particular context from scratch is – in practice – inefficient to a potentially dangerous degree.