This post proceeds from the orientation of this blog and from the claim advanced in the previous post: that science is not primarily a representational enterprise, but a practice that stabilises possibilities of instantiation.
If that claim is to do any real work, it must reconfigure what scientific objects are taken to be.
The quiet centrality of objects
Scientific discourse is saturated with objects.
Particles, fields, cells, organisms, signals, populations, systems — even when carefully qualified, these terms function grammatically and conceptually as things. They are treated as entities that exist, bear properties, and enter into relations.
This object-centred picture is rarely argued for. Like representation, it is inherited.
And like representation, it explains very little about how science actually functions.
Why objects do not do the work
Treating scientific entities as primitive objects creates a familiar set of puzzles:
Where, exactly, does one object end and another begin?
Which properties are essential, and which are contingent?
How can the same object appear differently under different experimental conditions?
How can new kinds of objects appear at all?
These puzzles are typically managed rather than resolved — by refining definitions, introducing hidden variables, or multiplying levels of description.
But the persistence of these problems is not accidental. They arise because objecthood is being asked to do explanatory work it cannot do.
Objects are outcomes of scientific practice, not its starting point.
Systems before things
What scientific work requires first is not an object, but a system.
Not a system understood as a collection of interacting things, but a system understood as a structured space of possibility.
A scientific system specifies:
what kinds of distinctions can be made,
what kinds of variation are admissible,
what constraints govern possible outcomes,
and which actualisations count as instances of the same phenomenon.
In other words, a system functions as a theory of its possible instances.
Only within such a system does it become meaningful to speak of objects at all.
Objects as perspectival actualisations
From within this frame, scientific objects are not ontological primitives.
They are perspectival actualisations within a system of constrained possibility.
What appears as an object is:
one way a system can be cut,
one configuration in which constraints are satisfied,
one stabilised pattern among many possible ones.
This is why the “same object” can behave differently under different conditions without contradiction: the object is not a thing with hidden properties, but an actualisation viewed from within a particular construal.
Object identity is not preserved by underlying substance, but by the stability of the system that makes certain actualisations count as the same.
Laws as constraints, not descriptions
This shift also recasts the status of scientific laws.
On an object-centred picture, laws describe how objects behave.
On a systems-as-potential picture, laws function as constraints on what can be actualised.
They do not tell us what must happen in every case; they delimit what can happen at all.
This is why laws tolerate exception, approximation, and breakdown without losing their force. Their role is not to catalogue behaviour, but to stabilise a space of possible instantiation.
What becomes visible
Once objects are treated as actualisations rather than primitives, several features of science become newly intelligible:
Why experimental setup matters more than observation.
Why individuation is always context-sensitive.
Why scientific change often involves redefining systems rather than discovering new things.
Why debates about “what really exists” so often stall.
Science does not begin with objects and then relate them.
It begins with relational constraints that make certain objects appear.
Preparing the next cut
If objects are outcomes of instantiation within structured systems, then experiments can no longer be understood as neutral probes of pre-existing things.
They are sites where systems are forced to actualise in particular ways.
The next post turns to this directly:
What does an experiment actually do?
No comments:
Post a Comment