Peter
It is neither arbitrary, not a belief system. Though do note that it is 'only' a generic statement.
As stated previously, but for convenience, I will repeat it here. The physical existence we are investigating is all that is potentially knowable to us (ie what we can be aware of because we can either experience it directly or can hypothesise it, which is, in effect, virtual sensing). Whether we can get to know all that is potentially available is another, secondary matter, the point is potentiality, as opposed to no potentiality. Knowability being enabled by a physical process, which involves the receipt of physical input. Which means that we are enabled to be aware of what may only be one particular form of existence, but as we cannot know (experience) an alternative this is irrelevant (hypothesis just being virtual sensing, ie it operates within the rules of sensing enabling us to state what we could have been able to sense had identifiable issues not prevented it. It is not a means of invoking which are beyond our existence). We cannot transcend our existence; we are trapped in an existentially closed system, which has as if determinant a physical process.
Now the key point here is that, therefore, physical existence is a confined, definitive circumstance. Which is why I can say I have no beliefs because my reference (context) is to physical existence as knowable to us, not some array of possible alternatives which we cannot know. It is meaningless to judge any comment against a possible alternative that we cannot know, just in the same way that it is not science to invoke such an alternative possibility. We can only attempt to explain the potentially knowable.
So, we know two key points about existence:
-it occurs independently of the mechanisms which enable its awareness, and occurrence must involve definitiveness
-comparison of inputs indicates that there is alteration, ie it occurs differently, but must be definitive in each case
This apparent dichotomy is resolved by sequence. Physical existence must be a sequence of definitive, discrete physically existent states of whatever comprises it. The important point to realise here is that objects do not exist as we conceptualise them. They are existent as a sequence of physically existent states, each one being reality at the time it occurs (there is no time in reality, only space). There just appears to be an object which persists over time, because we are defining object on the basis of superficial physical characteristics. We know there is alteration in objects, so just follow the logic through to its conclusion. Then think of sequence as 'tick'.
So, we cannot make pronouncements about occurrences outwith our existence (we have to accept the physical circumstance we are in) and the physical circumstance 'stops' with the interaction at the eye, it has nothing to do with any of the subsequent processing. I am not suggesting there is "no refraction by a lens", or whatever. This is just detail of no consequence in a generic definition. It is sufficient to say there is an interaction, in principle the same as that which occurs with a brick. The difference is that the eye/lens/whatever, is the front end of a system which can utilise the input rendered by the interaction, the brick cannot. Neither does it conflict with observation, again, in generic terms, it is sufficient to state that this involves the receipt of a physical input (known as light).
Rate of change does not only come from acceleration. Rate of change is the proper term for time, the frequency at which alteration is occurring (reality turnover rate). Again, for the generic explanation, forget all the detail about how light works, how it interrelates to what occurred, etc. The point is that there is a rate of alteration, a tick. Einstein used ageing, hand waving would have been better. Now, it does not matter if this rate is constant, slowing/increasing. It is the relative spatial relationship which matters (leaving other possible influences on light aside). Because if that spatial difference is altering, then the time taken for each 'tick', as represented by light, to reach the observer will alter. But the tick is whatever the tick is, in reality. It is not altering in the way that it appears to do so courtesy of an optical illusion. The fact (if true) that within 'each individual frame' (whatever that is) that the speed of light always calibrates to the same number, is irrelevant. What is relevant is the duration incurred from occurrence and creation of light, to receipt of light. During which time, the recipient wrt occurrence could have altered relative position. It is also irrelevant to the problem with relativity, because there is no observational light in relativity theory.
Paul