It seems that there are two extremes to the spectrum of interpretation.
At the simplest end is a simple dichotomy – true/false.
All of us must start from this most simple of systems, and most of us revert to it under stress.
At the other end of the spectrum is the purely probabilistic, awareness of there being many levels of uncertainty present in the many levels of processes leading to any perception.
The Oxford defines Fact as “Something that has really occurred or is actually the case”.
It seems that the more we learn about both the nature of reality, and the physical and logical processes of perception, the more uncertainties we become aware of.
Thus the idea of Fact is both invention and discovery.
A modern understanding of perception is that what we actually perceive as conscious entities isn’t reality itself, but is a subconsciously created model of reality.
In this sense, all perception is invention.
It is invention in that it is a model, not the thing itself.
That model seems to have been conditioned by multiple layers of heuristics selected by the differential survival of things over the deep times of biological and cultural evolution, and there can be personal experiential and conscious influences on the constructs present in the model.
Those models have utility to the degree that they have reliability in reality, that lead to survival.
It must also be discovery in several different senses:
In the sense of the levels of selection (both conscious and subconscious) that have taken place in assembling the components of the model; and
In the sense of the subconscious focus on that particular event as against any of the billions of other events one could have given focus to; and
In the sense of the degree of conscious attention one directs to the verification and substantiation of the probabilities involved in coming to a determination.
As well as our view on which contexts any particular fact has relevance.
For example – 3 different people, working on 3 different tasks at the same time and in the same town.
One is building a house, using the traditional carpentry tools of hammer, nails, saws, lumber, etc. For the carpenter, flat earth is a fact. He doesn’t worry about curvature, the floor is flat. What curvature might be present is below the errors of measurement in his instruments.
Nearby is a sailor, planning a trip across the Pacific to a small island. For him the earth is round, and a great circle route is the best path to try and stick to (provided winds and currents allow).
In the house next door is a software developer working on kernel code for a new set of satellites and receivers to provide GPS like positioning. That developer has to use an ovoid model of the earth and relativistic space-time coordinate systems and curved space if he is to achieve the sort of millimeter accuracy on positioning that he is aiming for,
Ovoid earth in curved space.
All three are true enough for their intended purposes, and all would probably fail in other contexts or for other purposes.
The more I learn about the levels of heuristics involved in the many levels of systems in brain that allow us to have the experience of being that we do, the more the boundary between truth and utility blurs.
There are so many levels at which our perceptual and conceptual systems are at base heuristics that worked well enough to survive, and that is as good as truth gets.
Once you start to simultaneously get a handle on quantum mechanics and deep neural networks (and their biochemical substrate), it really does get very blurred indeed.
Our systems evolved for macro-scale objects, not the substructure of our being.
We really are pushing things to their limits.
The complexity of reality seems to be so great that we have got to make vast simplifications to get to anything even vaguely approximating utility (knowledge, truth, whatever). The idea of ever getting a one to one correspondence between the the map and the territory is just so laughably implausible once you start to get a handle on the numbers and uncertainties present.
I like the way Jordan Peterson puts it – that we need to work to balance both order and chaos – excess or deficiency of either has serious problems. Truth for me sits in that boundary between order and chaos – held neither too firmly nor too lightly.
And I run a software company. I have been programming computers for 40+ years. I understand something of both the utility and the limits of logic and systems. I love David Snowden’s Cyefin Framework for the management of complexity, and Jo Ito’s principles (MIT Media Lab).
No Roland – you misinterpreted what I was trying to convey.
I started out as a biologist, interested in biochemistry and how brains work, and became interested in computers and all things algorithmic, logical and computational.
I am not talking about any sort of living in a simulation argument as Musk put it.
I am talking about the way our subconscious systems assemble information into a slightly predictive model of reality that seems to be the experiential reality of our consciousness.
I spent a lot of time on eyes and signal processing and transduction from eyes to brain in my undergrad studies 43 years ago. A lot of processing happens in the retina, then at many different sites within the brain simultaneously, all feeding back to the model.
And to me, as someone who has been programming computers for over 40 years, it is all easy to model conceptually, but I get it might not seem at all obvious to someone without a similar experience set.
Like Bruce says- there is a varying degree (context dependent) to which we see what we expect to see. That happens on every level we manage to instantiate.
Oh Yeah – when a computer starts doing unwanted stuff, a big mess can happen real quick (Poster on my wall – To err is human, to really foul things up requires a computer).
I use truth like I use a star when navigating a boat, as something in the far distance that I never expect to get to but acts as a useful direction to head in.
My destinations thus far all seem to be useful approximations in the contexts of their use, rather than any sort of absolute notion of Truth.
In a broad sense I agree Shaun, and it does seem to be a bit more complex than that.
It does seem to be a “fact” that the only information that we consciously get about reality is what makes it through the array of sensory, categorising and associative systems that subconsciously create the model of reality that is our conscious experiential reality.
So yes, there does seem to be a “reality” out there that does what it does, and it does in fact seem that the only access we as conscious entities have to it is via a set of systems that are mostly heuristically based, that have been selected by differential survival over genetic and cultural contexts, that are further modulated by our personal life histories.
Being as conscious as possible of the many layers of systems present in that model generating set of systems is fairly important if one is interested in generating as accurate a model as possible, both in the sense of useful correlation to what actually is in reality, and in terms of projections of possible future paths that we may influence the possibilities of instantiation through our conscious action.
Yudkowski’s Rationality from AI to Zombies looks at some aspects of the issues present, Jordan Peterson looks at a different set, Werner Heisenberg a different set again. All are important parts of a necessary understanding.
And one needs to get comfortable with the many levels of systems that are not predictable, be they maximally computationally complex (as per Wolfram et al), or chaotic, or complex adaptive systems (as per Snowden et al), or fully stochastic, or any other set of non-predictable systems (which may actually be an infinite class of sets).
It pays to be really comfortable with uncertainty and to plan with multiple sets of safety strategies, and not to get too confident about anything (and that seems to be always something of an art, every bit as much as it is science).
I used the term “usefulness” in the sense of adding to the accuracy of the model in use in the context under consideration.
There seem to be two major domains to which people apply the term “truth”.
One domain is the domain of experience, what we normally call “reality”. The evidence from modern science is that all information from this domain is probabilistic in at least two very distinct ways. It seems to be probabilistic at the quantum level, and it seems also to be probabilistic in terms of the many levels of measurement error and uncertainty that result from the tools and methods used. Using large data sets can reduce, but never eliminate, such uncertainty.
The other domain is the domain of abstract systems and logic, and even there Goedel has demonstrated necessary levels of incompleteness, that make the use of the modifier “absolute” questionable.
Given both of those domains, I prefer to use the term “useful heuristics” which tends to be explicit about the (many levels of) uncertainties present, while at the same time expressing the idea that such things can be exceptionally useful and reliable within particular contexts.
It also gives a hint that we can never be entirely certain about exactly where the boundaries of that confidence lie.
And the little von Neumann processor etched in silicon that resides at the heart of this laptop is exceptionally reliable, provided that I keep it within the ranges of temperature, humidity, source voltage, electric and magnetic field flux, vibration, acceleration, etc, that it is designed to operate within. Within those limits it works very reliably, beyond them not so much (or at all).
For me it is more like:
There is reality, whatever it is, and it seems to have aspects of both order and chaos in balance and tension, and to be fundamentally unknowable in many essential aspects.
Then there are our perceptions of reality, which seem to be more strictly speaking perceptions of a subconsciously created model of reality based on many levels of heuristics.
Then there are our conscious level models of reality. And no model is ever the thing it models. Models are only ever approximations to the thing being modeled at some level.
I acknowledge the power of maths and logic, and use them far more than most. And I also acknowledge that they are modelling tools, useful and not necessarily with strict correlates in reality. To get a flavour for what I mean by that, consider a circle. The ratio of the diameter and circumference of a circle is irrational (has no possible integer relationship). It seems that the reality we find ourselves in is quantised. Thus there seems to be no possibility of instantiating a perfect circle in reality. Thus the notion of circle is a good tool for approximating reality, and not an accurate description of reality.
Same goes for many aspects of logic.
It seems to me that the notion of truth as being any sort of absolute must be rejected in favour of a clear distinction between model and the thing being modeled – something more like an approximation or a heuristic.
Hence my reluctance to use the term “truth”, as its classical meaning is based in a set of assumptions that do not appear to be supported by the evidence.