Through recent discussions, I’ve found myself wanting to clarify where my own sympathies lie.
I find myself strongly resonating with the view associated with Merleau-Ponty —
the idea that we cannot be certain that an objective world exists as a fully completed structure,
entirely independent of observation or engagement.
This is not a denial of the world’s existence.
Rather, it is a refusal to take for granted the assumption that the world is given to us
as a finished object, already complete before any encounter with it.
We are not beings who apprehend the world from a completely detached, external standpoint.
We are embodied, acting, perceiving beings who are always already involved with it —
through movement, observation, and interaction.
In that sense, objectivity seems less like something guaranteed prior to experience,
and more like something that gradually stabilizes through engagement, sharing, and repetition.
This is not the claim that “everything is subjective.”
It is simply the sense that we do not need to presuppose a
purely observer-independent, unquestionably objective world
in order to think meaningfully about reality at all.
Thank you very much for this thoughtful comment. I think your formulation of the infinite regress problem is exactly right, and it is one of the main reasons why I have never been satisfied with accounts that treat observation as an active process that generates facts.
In fact, the very regress you describe is explicitly addressed in a paper I shared recently (the one dealing with absolute subjectivity and generated observers). One of its central claims is that as long as observation is treated as a physical interaction that produces facts, the theory is inevitably driven—exactly as you describe—either toward an arbitrary stopping point or toward an infinite regress in which observers must themselves be observed.
The approach taken in that paper is to shift the locus of fact-formation entirely away from observation itself. Observation is treated as fundamentally passive, while the stabilization of facts occurs not through observational acts, but at the level of relational structure itself—through decoherence and global constraints. In other words, the regress is not halted by positing a privileged observer, but dissolved by rejecting the assumption that observation is what performs the ontological work in the first place.
For this reason, I found your emphasis on the distinction between observer-dependence and contextuality especially resonant. The paper argues for almost exactly the same point: reality is thoroughly contextual, but not observer-dependent in an anthropomorphic sense. What we call an “observer” is itself a product of stabilized relational structure, not its origin.
As a side note, I have already shared the first and second papers of this series with you in replies to other posts.
The work as a whole consists of six papers, and given the way you have framed the infinite regress issue here, I think the fifth paper is the most directly relevant to your current comment. I would therefore like to share this fifth paper with you now.
After reading it, I would be very interested to hear whether you think this approach genuinely succeeds in avoiding the infinite regress you describe, or whether it merely relocates the problem elsewhere.
(Link to the paper) https://www.researchgate.net/publication/399613726_Non-transitive_Correlation_Structure_among_EEG_Brain_Topology_and_Quantum_Computation_A_Time-Series_Analysis_of_Subjectivity_Alignment_Conditions