Control is what control systems do. What a control system does is hold some measurable property of a thing at or close to some reference value, in spite of other influences that would change it.
A thermostat controls the temperature of something: a room, a fridge, a hot water tank.
A person standing up is controlling their standing posture.
The driver of a vehicle is controlling many things: the vehicle’s lateral position on the road, its speed, not hitting anything, following an intended route to an intended destination.
Control systems keep your heart beating and lungs breathing at a suitable rate to supply oxygen sufficient for the body’s needs, however those needs change.
Measuring this against your initial putative definitions:
Control is when an element makes another element do something. This relies on elements “doing stuff”.
Ice freezing in cracks will split a rock. Is the ice “controlling” the rock? No, this is not an example of control.
Part of the traditional process of manufacturing slate tiles is to expose quarried slate slabs to water in low winter temperatures in order that the freezing water will split it into suitably thin sheets. The slateworkers doing this are engaging in control: they act as necessary to get the slate into the desired form. The water that they are using is not controlling anything.
Control is when an element {counterfactually, evidentially, causally, logically...} determines {the behavior, the outcome of the behavior} of an assembly of elements.
That is only causation, not control. An avalanche will determine the behaviour of the village it is thundering towards: it will obliterate it. This is not control. The avalanche would act no differently if there were no village there, nor if the buildings were strong enough not to be swept away. If the village happens only to lie near the path of the avalanche, the avalanche will not notice that it is going to miss and veer towards it.
Control is when an element modifies the state of an element. This relies on elements having a state. Alternatively, control is when an element replaces an element with a similar element.
Again, many things modify the state of many other things. This also does not capture the idea of control: the one thing bringing the other thing into a specific state and keeping it there, the thing varying its actions so as to produce that result.
Control is when an element selects something according to a criterion.
This contains only a fragment of the idea of control. To have a control system, the thing selected must be an action and the criterion is that that action must bring the controlled variable closer to its reference value.
For control to be present, these things must be present.
The variable to be controlled.
A means of sensing that variable.
A reference value for the variable.
A range of available actions that will influence that variable.
A rule to determine an action, given the reference value and the sensed value, such that the selected action always tends to reduce the difference between them.
Identifying all of these things in the case of the thermostat is left as an exercise. In this example, all five are visible, physical things that you can point at.
I suggest that this is precisely the concept that you are searching for.
I suggest that this is precisely the concept that you are searching for.
No.
The point of talking about control is to lay part of the groundwork for understanding what determines what directions a mind ends up pushing the world in.
This notion of control is fine but doesn’t answer the question.
what determines what directions a mind ends up pushing the world in.
What determines what directions a mind ends up pushing the world in is that mind’s purposes — the reference values of its control systems. These determine the narrow class of states towards which it herds the world state, at all levels from seeking water to quench thirst, to striving for grand political ideals.
Sure, if the mind really is some control systems and you know how it’s some control systems and where their reference points are and how they combine. You don’t know that so it’s not a good enough answer, and I don’t see a good reason to restrict to this one model-piece, in trying to find ideas that could provide a good enough answer.
There are many theories of what the mind is and how it works, and nobody knows which if any of them are right. I have presented one idea (not originating with me, BTW), and I do not see a reason in what you have said to exclude it from the ideas that could provide a good enough answer.
But the question, “what determines what directions a mind ends up pushing the world in” is somewhat vague, and perhaps you are looking for a completely different sort of thing than a theory of how the mind works.
The idea of control system shouldn’t be excluded, but it’s far from answering the question. Even if the answer does look like “it’s a control system”, we still don’t know HOW minds (e.g. human minds) are control systems—we’d want to be able to look at a mind and then say “oh, see, it’s a control system, here’s the action-selector and here’s the reference value and here’s the comparator” and so on. And we would still want to understand whence comes the reference point, and whence comes the language in which the reference point is stated. Though I’m skeptical that a satisfying answer could look like that.
Perhaps I should have given more of the background. Hang on, I did. The theory is not that the mind “is a control system”, but that it and the body are made of control systems, very large numbers of them, arranged in a particular hierarchical way, the output of higher-level systems providing the references for those in the layer below. At the foot of the hierarchy is muscular control and biochemical control loops. The lower levels of this are standard physiology and biochemistry. The upper levels are more speculative, and there is as yet not even conjectures about the morphogenesis of the whole hierarchy, i.e. how it is built during development. (For that matter, not much is known about physical morphogenesis either.)
It seems to me that TsviBT is trying to figure out for good which of them is definitely right in terms of the concept of what control is in all cases and contexts.
Yet he objects that I do not know if what I described is definitely right. No-one knows what is definitely right. If he wants to find out, all that other people can contribute is possible ideas for him to measure against whatever the idea in his mind is. Well, I have given one. What he does with it is up to him.
I do.
Control is what control systems do. What a control system does is hold some measurable property of a thing at or close to some reference value, in spite of other influences that would change it.
A thermostat controls the temperature of something: a room, a fridge, a hot water tank.
A person standing up is controlling their standing posture.
The driver of a vehicle is controlling many things: the vehicle’s lateral position on the road, its speed, not hitting anything, following an intended route to an intended destination.
Control systems keep your heart beating and lungs breathing at a suitable rate to supply oxygen sufficient for the body’s needs, however those needs change.
Measuring this against your initial putative definitions:
Ice freezing in cracks will split a rock. Is the ice “controlling” the rock? No, this is not an example of control.
Part of the traditional process of manufacturing slate tiles is to expose quarried slate slabs to water in low winter temperatures in order that the freezing water will split it into suitably thin sheets. The slateworkers doing this are engaging in control: they act as necessary to get the slate into the desired form. The water that they are using is not controlling anything.
That is only causation, not control. An avalanche will determine the behaviour of the village it is thundering towards: it will obliterate it. This is not control. The avalanche would act no differently if there were no village there, nor if the buildings were strong enough not to be swept away. If the village happens only to lie near the path of the avalanche, the avalanche will not notice that it is going to miss and veer towards it.
Again, many things modify the state of many other things. This also does not capture the idea of control: the one thing bringing the other thing into a specific state and keeping it there, the thing varying its actions so as to produce that result.
This contains only a fragment of the idea of control. To have a control system, the thing selected must be an action and the criterion is that that action must bring the controlled variable closer to its reference value.
For control to be present, these things must be present.
The variable to be controlled.
A means of sensing that variable.
A reference value for the variable.
A range of available actions that will influence that variable.
A rule to determine an action, given the reference value and the sensed value, such that the selected action always tends to reduce the difference between them.
Identifying all of these things in the case of the thermostat is left as an exercise. In this example, all five are visible, physical things that you can point at.
I suggest that this is precisely the concept that you are searching for.
No.
This notion of control is fine but doesn’t answer the question.
What determines what directions a mind ends up pushing the world in is that mind’s purposes — the reference values of its control systems. These determine the narrow class of states towards which it herds the world state, at all levels from seeking water to quench thirst, to striving for grand political ideals.
Sure, if the mind really is some control systems and you know how it’s some control systems and where their reference points are and how they combine. You don’t know that so it’s not a good enough answer, and I don’t see a good reason to restrict to this one model-piece, in trying to find ideas that could provide a good enough answer.
There are many theories of what the mind is and how it works, and nobody knows which if any of them are right. I have presented one idea (not originating with me, BTW), and I do not see a reason in what you have said to exclude it from the ideas that could provide a good enough answer.
But the question, “what determines what directions a mind ends up pushing the world in” is somewhat vague, and perhaps you are looking for a completely different sort of thing than a theory of how the mind works.
The idea of control system shouldn’t be excluded, but it’s far from answering the question. Even if the answer does look like “it’s a control system”, we still don’t know HOW minds (e.g. human minds) are control systems—we’d want to be able to look at a mind and then say “oh, see, it’s a control system, here’s the action-selector and here’s the reference value and here’s the comparator” and so on. And we would still want to understand whence comes the reference point, and whence comes the language in which the reference point is stated. Though I’m skeptical that a satisfying answer could look like that.
Perhaps I should have given more of the background. Hang on, I did. The theory is not that the mind “is a control system”, but that it and the body are made of control systems, very large numbers of them, arranged in a particular hierarchical way, the output of higher-level systems providing the references for those in the layer below. At the foot of the hierarchy is muscular control and biochemical control loops. The lower levels of this are standard physiology and biochemistry. The upper levels are more speculative, and there is as yet not even conjectures about the morphogenesis of the whole hierarchy, i.e. how it is built during development. (For that matter, not much is known about physical morphogenesis either.)
So there it is: consider it, or not.
It seems to me that TsviBT is trying to figure out for good which of them is definitely right in terms of the concept of what control is in all cases and contexts.
Yet he objects that I do not know if what I described is definitely right. No-one knows what is definitely right. If he wants to find out, all that other people can contribute is possible ideas for him to measure against whatever the idea in his mind is. Well, I have given one. What he does with it is up to him.