Knowledge

My Positions

1. I consider knowledge to be a mental model, held in the neurological structure of the brain, built from sense data (experience) and shaped by intuitions that approximate reasoning mechanisms such as deductive, inductive, abductive, and analogical reasoning. The model is constructed to enable prediction, with the aim of informing action and minimising surprise.
2. I take as a necessary presupposition that there is an external reality which is the origin of my sense data, and that this external reality remains consistent, in that it continues to behave in the manner described by the laws of logic and physics.
3. I recognise that sense data and reasoning tools are fallible, and that humans do not have direct access to external reality.
4. Knowledge may be considered sufficiently true, or sufficiently accurate, insofar as its predictions remain uncontradicted by future sense data or reasoning.
5. A component of knowledge is the anticipated likelihood that a model will be sufficient to avoid surprise. This could be called one's confidence in a model. It allows provisional, low-confidence models to suffice until new data or reasoning can be applied.

Notes

1. I do not wish to mistake the map for the territory. I do not consider knowledge to be True, only presently sufficiently true and uncontested (Fallibilism).
2. Karl Friston's free energy principle is especially relevant here. On this view the brain is a prediction engine, and perception, learning, and action are all strategies for minimising the gap between what the model predicts and what the senses report.
3. It follows that to change what someone considers to be knowledge, we must introduce surprise: a prediction of their model must visibly fail.

1. If we are cast as an enemy, nothing we say will be sufficiently surprising, because it is the nature of an enemy to try to surprise you, and so you are not surprised. This is why hostile debate rarely changes minds, and why approaches such as Street Epistemology try to get people to test their own models rather than defend them.
   4. The instinct to reduce surprise functionally maps reality, insofar as reality is the assumed source of the surprise. It has the practical advantage that you no longer need to solve the entire nature of reality before deciding whether or not to get up in the morning.
   5. This account also explains why intuition feels like knowledge. Intuition is a fast, compiled model. It is often right within the environment it was trained in, and often wrong outside it (Think horses, not zebras in one context becomes a Hasty Generalisation in another).
   6. Confidence should be proportioned to the evidence (Empiricist principle). A model that has survived many attempts at refutation earns higher confidence than one that has not been tested.

Also see

1. Justified True Belief
2. True
3. Certainty
4. Data and Evidence
5. Alfred Korzybski