Kristoffer Balintona — Notes

Obsession (2026) is good horror

Thu, 23 Jul 2026 22:16:00 GMT

I was urged by two close friends to watch this movie, specifically in theaters. I normally don’t watch horror, but these two normally don’t recommend movies or shows—so I had to check it out. Below are some of my thoughts after watching.

Immediate impression §

This was a really good movie. Horror movies with inventive ways to sustain tension are really rare to come by. And this one was a brilliant orchestration.

On first thought I’d say the premise was unique and centrally compelling. But thinking about it, that isn’t quite right: the premise is itself not so remarkable. Really, it’s just: we overshoot our wishes, we seek to go where we actually don’t want to be. What’s so compelling about Obsession is more its self-coherence, the selection of elements to constitute a perfect storm.

This is the shifty trait I see as intrinsic to exceptional works of art.

In the context of this movie (and its genre: horror), it’s how it “sustains tension” so effortlessly, a series of moments spontaneous enough to present themselves as merely the contingent unfolding of events. Of course the whole movie is orchestrated—but it’s precisely the motion of naturalness that backgrounds the viewer’s present act of watching and foregrounds the moments that unfold as moments (not as “orchestrated moments in a movie”). A kind of self-propulsion of the inevitable. The phenomenology is: to take spectating out from the spectacle and to take up (occupy) the bodies on the screen.

Reflections §

You know, I was talking with Aaron and realized something unique/rare about the movie. It’s the closeness of Nikki. She’s kept close to Bear—too close (which references the title and probably one of the main symbolic messages of the film). It creates a disturbing sense of claustrophobia, I’d say, for the viewer: the source of horror looms over your shoulder, at the back of your head, and an unceasing threat—and in the movie, it’s precisely manifest as those moments where Nikki stares hidden from a corner, when she stands motionless inside the door frame, when she cuddles in bed with Bear; then cashed in and realized in other moments by the opposite: startling movement, say, when she dashes into view and bashes Sarah’s head in.

Compare this to the typical horror movie. Almost always the source of horror starts distant and gets closer (in the “felt,” phenomenological sense, but also sometimes literally). It’s the cookie cutter way to create scenarios where directors can have an excuse insert the typical jumpscare or “running away to safety”- or “readying to confront and battle the monster”-type scene.^[1] But because of this, the rhythm of the horror becomes predictable and template: the viewer can feel the rising and receding action, alongside the climaxes and dips. On the other hand, Obsession works by having the source of horror felt constantly.

But Obssession is ingenious here: the reason the typical horror flick doesn’t try to do the same thing is because it can’t. If you just place the “horrifying monster” around the corner or in the room of every scene, then the shock value of the monster disappears; it’s no longer scary. In those movies, the monster is horrifying precisely because it isn’t seen. So usually the source of horror can’t just always be seen. Obsession gets away with it because of its premise + selection of elements. Nikki is a girlfriend who wants to stay close to Bear, and Bear wants to stay close to Nikki. Naturally, Nikki stays over at Bear’s place and sleeps in his bed with him. He goes to work (with lunch she’s prepared) and returns home to her. He wakes up in the morning with her having prepared breakfast or reading. He wakes up during the night and she’s in the corner of the room staring at her. She’s always there, and the claustrophobic proximity is disturbing. (Even when she’s “actually scary and deranged” and not “trying to be normal but terrifying,” her physical proximity to the character is uncomfortable.)

Speaking of which, I should amend one of my opening remarks, taking it further: I said Nikki keeps close to Bear, but I think it’s better to say Nikki is kept close by Bear. Of course Nikki’s possession or curse (or whatever you want to call it) means she’s always trying to be with Bear, but Bear is the one who allows her to keep close (and, it goes without saying, Bear is the one whose wish caused all this). That’s the kicker and tension in the movie: Bear isn’t running from the horror, his attraction to it (Nikki) is what makes it scary, it’s the source of the tension. (And, of course, the philosophical rub is that Bear had many chances to try backing out of the deal or even undo its harm but he decides not to—he knows something is wrong from the beginning, he rejects the lending hands of those in his life, he doesn’t share what he did to his friends so that they could brainstorm solutions together since he doesn’t want to take accountability for what he did (or wanted), he learns that his suicide can save real Nikki but doesn’t turn to that option until the very end, and when he does he fails to build to courage to shoot himself, and when he opts instead to go out the gentle way, by overdosing on pills, he chickens out and changes his mind^[2].) The viewer wants to get away from Nikki and out of the situation but is forced to embrace Bear’s desire, watching the avoidable disaster live, to try and scrounge a relationship Nikki, the source for horror. That proximity to what shouldn’t be proximal is how the movie naturally generates tension in almost every scene.

Speaking of Nikki: I think her character as the source of the horror is also interesting. It’s unlike the typical horror monster. Usually, the monster is either transparently evil—it chases you, wants to harm you, wants to attack your friends and family, and so on—or it’s horrifying by being mysterious, cryptic, and creepy. Nikki is neither: she’s sentient and wants to do good (in combination with, of course, forceful violence). Of course, she doesn’t do good—but it’s the fact that she has aims to please: it’s just that it’s a sick, twisted version of good; she’s so single-minded and infatuated that what she does edges into the demented.

Guile Exceptions

Thu, 23 Jul 2026 01:42:00 GMT

Learning Guile has been on my wishlist for a while because of the power it affords users of Guix: you get to control its service manager, Shepherd^[1], configure other programs, manage containers^[2], all in the same powerful language in a consistent syntax. But Guile is also a power language all on its own, too. You can write standalone program with it happily (and integrate it with your Guix config seamlessly).

I recently found my first excuse to dive into the language itself (not just for Guix). My idea was to have a small, personal web server with a RESTful API^[3] that lets me run, stop, and see the status and output of pre-defined tasks. I read the relevant pages in Guile’s manual many times and within a few days got something that was acceptable (though I’ll no doubt refine, refactor, and add to it as time passes).

One of the cool things I encountered with Guile were its exceptions. I’m not a programmer by trade and haven’t used many languages before, so I was mostly only familiar with “error = program implodes.” I found out Guile obviously has error, but it also has quite rich exceptions. The older system was just try + catch, but the more modern 2019+ set of features is based on records (I think), has hierarchies, and more powerful “catching.”

It was quite hard to wrap my mind around it enough to even just get off the ground, but I found these two articles very helpful, alongside the Guile manual pages on exceptions:

Civil War (2024) was incredible

Sat, 27 Jun 2026 03:14:00 GMT

Earlier this week I watched Civil War (2024), directed and written by Alex Garland. I wanted to watch this movie since I noticed it on Alex Garland’s resume, Alex Garland being the director, writer, or screenplay writer of movies and shows including:

Ex Machina (2014),
Devs (2020),
Annihilation (2018),
Sunshine (2007),
28 Days Later (2002), and
28 Years Later (2025).

These are all motion pictures that I either like, really like, or are among my all time favorites^[1]. Naturally, I need to check this movie out.

Below is the unedited form of my post-watch writing; they present my thinking as it unfolded. They aren’t all the thoughts I have, but they are the ones I had during and immediately after watching.

Mid-movie impressions §

40 min in: I already love this movie. Love the brutal tone they went for. And love the concept: we’re seeing on-the-ground journalism of our own country during a hypothetical civil war. It is a forecast of where we could be if we’re not careful.

It reckons. It forces us to think about how things would be—then realize that in a way things are already like it (just not as bad). And in that way, the viewer is re-thrown into the present equipped with a new way to decipher it that was always already there.

Immediate post-watch impressions §

WOW. An incredible film. Almost speechless. It was perfectly, superbly candid. Exposing.

What a feeling to see what in real life are the forces meant to defend our leaders and country be attacked by the “other forces” of our country (secessionist forces). To see the White House and Lincoln Memorial bombarded. The rawness of the final act in the White House…

And that ending—gives one a ton to chew on. What was the Jessie feeling? What was Joel feeling? Why didn’t he mourn, or lash out like he did when Sammy (the older journalist) died? What did he say to the Jessie? Why did the girl take photos of Lee? Will she delete them? What does it mean when she does or doesn’t, especially in light of the fact that Lee deleted the photos of Sammy’s corpse? What were the soldiers and Joel thinking when they looked back at Jessie taking the photos?

I really did think, up until the ending, Jessie was the one that was going to die. And that this was a story that digests Jessie’s death. I was surprised when Lee was the one who died, though I thought that was the second (and only other, really, especially to save Jessie) most likely option. And naively, I thought they would get an interview with the President somehow—really fitting that all that we got was a primal, harrowing, desperate plea to survive, and that that the journalists (specifically Joel, who we saw early on in the car was the one they planned to do the talking^[2] to the President) knew that this was enough. War isn’t glamorous and doesn’t wait; so neither did the soldiers nor the journalists. (And, thinking about it now, that all that they had done, sacrificed, risked, and experienced up till then was in service for that singular, final sentence. Final words are part of what makes this film’s portrayal so raw: Sammy had no final speech, nor did Lee, and we don’t even get to hear what Joel and Jessie said or thought after Lee’s death.)

The movie did one thing exceedingly well: evoke emotion. Which is an exceptionally well-crafted experience.

In a way, this is journalism on journalism. To see the behind-the-scenes of the people who give us a look into the behind-the-scenes.

Further thoughts §

One §

Thinking about it now and reading a bit of others’ thoughts on Reddit for a few minutes I now see: Lee “had no life left in her.” In the sense of: she didn’t have the will to continue on being a journalist. She herself, in her professional advice to Jessie, said that to be a professional photojournalist you have to distance yourself from the horror, you have to accept that you’re there to record history, not change its course.

But that’s precisely what we see absent in Lee through the film.

Thinking back, we get hints of this from the very beginning: she cares for Jessie. She saves her from danger in the opening scene. She protests against her joining their group (though she inevitably accepts, though I think that came more from her outward identity as a professional journalist; her interior struggle was the source of protest). Then, I generally didn’t see Lee taking many photos “in the action.” IIRC, almost all the instances where I noticed Jessie taking photos were outside of action, like in the car. She didn’t really take any photos midst brutality. And the biggest hint to the viewer of all this is that she deletes the photo of Sammy’s corpse. I think earlier she implied to Jessie that yes, if she did get shot she would in fact take a photo of her body. All indications of Lee drifting from the distance and ruthlessness, in a way, of her photojournalist passion. Even in the small, quaint town they find, she entertains the idea of wearing then buying a dress, something she outwardly (her identity as a journalists) protests despite her (interior uncertainty) eventually seceding.

This peaks in the final act where the group goes into D.C. Lee can’t hold herself together. During gunfights she just stands there or breaks down in a panic attack. She can’t handle it. Only in the quieter moments, does her cunningness as professional photojournalist come out: she identifies that the lead they want is in the White House, before even the soldiers realize it.

And, of course, the very act of saving Jessie epitomizes the fact that Jessie no longer has what kept her going as a journalist. Because of the prior dialogue with Jessie, where it’s implied that she would take a photo of Jessie’s body getting shot, and the fact that Lee deleted photo’s of Sammy’s dead body, we should know that Lee-as-a-journalist would’ve let Jessie die, capturing that “glorious” shot. But she dives into the line of fire to save Jessie.

Lee was done as a journalist. Her experiences through wars and chaos broke her. And it’s also precisely why Jessie (and even Joel, I think) could act cold in the face of her death: they weren’t done as journalists. They could temporarily steel themselves and shrug it off for the sake of capturing the moment. It’s why Jessie’s first instinct is to take photos of Lee getting shot—which is precisely the advice Lee gave to her in order for her to be a good journalist—and why Joel marches forward into the Oval Office to catch the President’s last words before the soldiers execute.

So for this reason: I think Jessie keeps the photos of Lee’s death (and probably why we, as the viewer, see those photos taken to begin with, as opposed to us never seeing Lee take the photo of Sammy that she deleted).

Maybe this was a story about the human toll it journalism demands, in the form of the passing on of the torch from Lee to Jessie.

Two §

I also now appreciate that Lee’s “breaking point” was Sammy’s death. After that is truly when she’s almost incapable of doing the duty of a photojournalist. It’s when she starts blankly staring and standing midst the war with an empty look (while Jessie and Joel still get off on the thrill of the hunt, so to speak).

And recall what Lee says to Jessie regarding Sammy’s death: “He could’ve gone out in many more worse ways” (roughly put). The same is exactly true of Lee: she could’ve gone out in many other, worse ways (during the film and of course before): she was diving into lines of fire during the final act. It’s when she’s confronted with Jessie’s possible death does she commit to the “non-journalistic” leap: don’t document Jessie, save her.

Her confrontation with the sum total of Sammy’s life, the life of a close friend and colleague, is what throws her to contemplate the same about herself. In so many words, I think this is what Lee underwent in that moment before sacrifice.

Great article on “Emacs introspection” basics

Wed, 24 Jun 2026 17:43:00 GMT

I recently read a great article for Emacs newbies: Emacs Teaches Emacs: The Missing README from Charlie Holland’s blog. In my opinion, a contender for the most important thing newbies should know when they start using Emacs (the step 0 to understanding the “value proposition” of Emacs, if you will) is its ability to introspect. That is, the commands and key sequences that let users learn about new parts of Emacs, understand how Emacs works, discover user options and commands, and do all of the above for arbitrary Elisp source code (e.g., from third-party packages).

Take a peak at the table of contents:

Two Keys
Start with M-x
The Help Commands
The Prefix Trick
1. find-
2. xref-
3. Info-/info-
4. apropos-
5. customize-
C-h C-h Help Cheatsheet
More Magic with which-key
The Coda

The article includes probably every tip I’d mention to an Emacs beginner.

First time web dev

Tue, 23 Jun 2026 09:34:00 GMT

My first touch with actual web development has been the development of this site using Astro. Previously, I used Hugo to generate the predecessor to the current site’s design—but I don’t consider that “real” web development since much of the logic and hard work has already been done for me (by the static site generator).

Overall, it has been quite pleasant, relative to how large of a meal Astro + Node + HTML + CSS + routing + deployment + more is.

HTML tag semantics §

HTML tag semantics matter. Don’t be confused by the fact that, by default, tags have styling. How a page looks should be controlled by CSS.
- Don’t confuse the default look of elements for their role in the document
- It is intuitive to use <i>, <b>, <br>, <em>, and so on for basic styling. However, HTML tags should convey the meaning of text, not what it looks like.
- In practice, this means these tags should be used pretty much only in prose, not in UI elements like the footer, navbar, site header, and so on.
HTML semantics also matter for contexts in which your HTML is presented outside of your pages. For example, reader modes (e.g., Firefox’s Reader View), or RSS feeds: these are contexts in which the styling of your text is inferred from your HTML tags and attributes, not your stylesheets.
Semantics for a few key or rare (but useful) tags:
<body>
The “actual” (displayed, rendered) content of the page.
<main>
Dominant content of the page. Typically includes everything but the site footer and header. Only one per page.
<article>
Represents a self-contained piece of composition. Can be used multiple times a page. For example, a blog post or each card in an article listing.
<header>
Introductory content (to site, to article, etc.). Can be used multiple times a page.
<nav>
Provides navigational links.
<footer>
Closing and supplementary information (e.g., author email address, copyright license, related links) about a given section (e.g., article). Can be used multiple times on a page.
<section>
Delimit thematic grouping of elements in a page. Most of the time, these involves a heading element.
<time>
Represent a period or point in time. Useful to search engines and other site features that use machine-readable time formats.

Accessibility §

Don’t forget about it! Consider theme contrast, non-sighted users, and user preferences like preferences for high-contrast (prefers-contrast) or no motions (prefers-reduced-motion).
Know ARIA standards and attributes, like aria-label, aria-labelledby, aria-pressed, and WCAG recommendations.
There are conventions for accessible use of heading tags. How-to: Accessible heading structure - The A11Y Project explains, alongside sharing a few browser add-ons to help.

Select one or two fonts §

Fewer is better; aim for simplicity and elegance. Almost all sites should only really need 2 fonts. Currently, this site has a main “for prose” font (serif) and a secondary font (sans serif) for various other places, e.g. UI elements and heading text.
Real life examples: Fonts In Use - Fonts In Use, phenomenal source to see what fonts people have used with each other
Font providers
- Fontsource
- Browse Fonts - Google Fonts

Fluid/responsive typography and spacing §

See Modern Fluid Typography Using CSS Clamp — Smashing Magazine and Generating `font-size` CSS Rules and Creating a Fluid Type Scale | Modern CSS Solutions for a deep dive on the matter.
The generator I’ve used for this site: Fluid Responsive Design | Utopia.
- I also use it to generate “fluid spacing”: CSS variables for spacing that are responsive.
The general rationale behind appropriate spacing in design and UI: The Secret Science of Perfect Spacing - YouTube.

There’s MD, MDX, and Markdoc §

I had to make a decision on what file format my content should be in.
I chose MDX, since I think it is as expressive as possible. (I export from Org-mode to MDX and don’t think there’s any HTML I want to emit that I can’t have my backend export into MDX.)
- Exporting to MDX makes for an easier time than exporting directly to HTML since MDX takes advantage of Astro components as well as opens the possibility of inserting Remark integrations to further customize the emitted HTML (although ideally that customization should be done by modifying my Ox-mdx exporter). I’d rather tidy up and standardize the final HTML output through Astro and its integrations, not Org-mode exporter stage.
But Markdoc was an enticing, albeit less common and support (by third-party packages), option. See Markdoc | Frequently asked questions and Markdoc | The Markdoc syntax.

Generators and aggregators §

Fluid typography (font sizing that changes appropriately based on viewport)
- https://utopia.fyi/
- https://typescale.com/
Content body width
- Baymard Institute
- Butterick’s Practical Typography, a well-regarded online typography book
Color palettes

Astro components are awesome §

They’re great. “Beginner friendly, simple, and powerful yet a thin abstraction” is my impression.
CSS selectors get scoped (but not rules, which are rendered normally).
Makes putting related things (e.g., CSS rules, HTML, JS) near one another.

Favicons §

Favicons and SVGs
- SVGOMG - SVGO’s Missing GUI for minifying SVGs, clean up SVG markup
- SVGcode—Convert raster images to SVG vector graphics, convert images into a SVGs

Don’t forget to have a favicon for each of your site’s color themes. You can have a different SVG for each theme, or have a style block in a single SVG which uses CSS media queries to change the colors of the SVG dynamically. For example, this site currently uses something like:

<!-- Adaptive-color favicon, depending on `prefers-color-scheme` -->
<svg xmlns="http://www.w3.org/2000/svg" viewBox="...">
  <style>
    .icon-fill {
        fill: #1a1a1a;
    }

    @media (prefers-color-scheme: dark) {
        .icon-fill {
            fill: #ffffff;
        }
    }
  </style>
  <path class="icon-fill" ... />
  <path class="icon-fill" ... />
  <path class="icon-fill" ... />
</svg>

Learned about “The Harada Method”

Tue, 23 Jun 2026 07:43:00 GMT

I learned about The Harada Method today from a recent edition (June 20, 2026) of Liz Jung’s Flourished Hope newsletter. A quick Google^[1] search later and I learned more about the method from this article: Crush your goals the Ohtani way, by the Harvard Gazette.

The gist of it is basically this. You have a big, ambitious goal. One for years down the line. Imagine a 3x3 grid; place that goal in the center. Then think of 8 constituent goals or behaviors and place them in each of the surrounding 8 boxes in the grid. Then you can go further: create a 3x3 grid for each of the constituent goals and do the same for each. In the end you have a ton of small, doable, realistic steps to your ultimate aim. Apparently this was Shohei Ohtani’s^[2] own Harada Grid when he was in high school:

Pretty neat! I’ve had an interest in goal-setting for years. Basically, the Ancient Greek “telos.” (But I think more in terms of the phenomenological tradition.) I’ve long identified the necessity (a belief, or opinion, of mine, of course) of making clear one’s goals in the role of one’s ability to self-author their life. Though it might be inaccurate to say that we’re “making clear” of these goals, since I don’t think they are just sitting there pre-formed, then we come and realize “Oh, that’s what I wanted all along!” I think it’s probably more like they are formed in the process of discovery. (Again, I gravitate to the existential-phenomenological accounts. With a hint of Lacanian psychoanalysis.)

And at a higher level: an interest in the curiously elusive “What now—and why?”—and having pondered what my own is.

Anyway, The Harada Method is just another interesting way to “method-ize” goal-setting. To make a procedure out of something enigmatic. Or: a science out of something philosophical. It’s a useful framing and self-cue.

The Visible and the Invisible: Chapter 1 — Reflection and Interrogation

Thu, 19 Dec 2024 00:45:00 GMT

This post contains my writing on The Visible and the Invisible chapter one, “Reflection and Interrogation.” The headings here correspond to the subdivisions of chapter one. This writing was written without the intention for sharing. It was meant for me to express the ideas I had after directly reading the chapter. As such, the understanding shown below is merely a draft, so to speak.

The Perceptual Faith and Its Obscurity §

This section^[1] presupposes an understanding of the Phenomenology of Perception because its goal seems to merely be to elucidate the dilemma fundamental to MP’s metaphysics: how the subjective contacts the objective, or how our private worlds meet at a shared world (e.g. Merleau-Ponty 1968, 10–11). This follows from the earlier comments that philosophy is not about word-meanings—philosophy is not of the order of the said or written—but about bringing to expression the silence of things themselves (Merleau-Ponty 1968, 4). Hence, I’d say the motif of this section is to explain how there seems to be dichotomies or poles but that they are in actuality illusions of a single moment. I anticipate this to be the inquisitive groundwork for the book.

Science Presupposes the Perceptual Faith and Does Not Elucidate It §

As the title of the section suggests, MP explains how science portrays itself as revealing Being, when it in fact obscures it by covering up the mystery of perception (i.e. its perceptual faith). MP characterizes the ontology of science, physics, and psychology alike, as objectivist ontology, which leaves the objective inaccessible to the subjective (e.g. Merleau-Ponty 1968, 16)—physics takes the absolute viewpoint of a subject look out onto the objective world, while psychology assumes an absolute spectator that witnesses the subject trapped in subjectivity (Merleau-Ponty 1968, 19–20). It is the task of philosophy to abandon this subject-object ontology, which is something like an approximation of reality (Merleau-Ponty 1968, 23), by looking at the context of the unified, total experience the distinction emerges from. (MP mentions some conclusions about perception from the introduction of the Phenomenology of Perception here, namely, how the perceptual field cannot be fully explained/accounted for solely by composing or summing external/objective conditions, for we saw that when we limit ourselves to objective conditions, the conditioned conditions other conditions; hence, there are conditions, but they are not of the order of the objective (Merleau-Ponty 1968, 20–21).) Importantly, this does not mean showing that there is another “order of reality” (the mental over the material, or the interior over the exterior) outside what science’s objectivist ontology conceives, for that would be to somehow once again fall into a subject-object ontology (Merleau-Ponty 1968, 22), nor does it mean objective thought (which seems to be another way of saying objectivist ontology? Not sure) should just be abandoned as an incorrect fiction, for science is actually just a phase that should be overcome by means of this re-examination of subject-object ontology (Merleau-Ponty 1968, 24).

Perceptual Faith and Reflection §

Situate a person in this passage/text who has never read it before. MP is coming from explaining his gripes with how the mystery of perception has been covered over by that which attempts to explain the world. He now discusses how this all revolves around the failings of perception, and he anticipates the possibility of a philosophy of reflection evolving into a philosophy of hyper-reflection.
Communicate to this person the ideas I found most interesting or relevant from the passage/text. This section appears very foundational to understanding the crucial shifts in his metaphysical thinking from his earlier works (e.g. Phenomenology of Perception). It centers around describing (somewhat) what a philosophy of reflection is, where it goes wrong, and how we can move beyond it.

Upon first impression, I identified the term “philosophy of reflection” as synonymous to “intellectualism” as the term is used in Phenomenology of Perception. But quickly became obvious that this is a designation for a new category: I believe MP truly refers to all philosophizing that involves reflection, or analysis—so this would include both intellectualism and empiricism but also, I think for the most part, the reasoning that Phenomenology of Perception takes up.

Philosophies of reflection are philosophies that employ reflection. I think “reflection” refers to the looking-back at perceptions and analyzing them, breaking them down to discover their nature; for instance, Husserl’s goal was to discover the invariant essences of experiences (transcendental reduction) to arrive at the essence of the intentional object (eidetic reduction). But realist scientists also take themselves to be peering into the nature of reality by looking-then-analyzing it. I think basically all philosophy and science reflects. MP’s argument is that reflection necessarily misses the mark because its claims are retroactive, a “regressive analysis” he writes once (Merleau-Ponty 1968, 45): they are made about a perception, so what reflection yields is not the original perception per se but a “perception-reflection-upon,” and instead of a thing, the result is a “thing-perceived-within-perception-reflected-upon” (Merleau-Ponty 1968, 38). Philosophy of reflection thinks that the reflecting upon something straightforwardly breaks down what that thing is, when in reality, the initial perception prior to the understanding of analysis stands untouched because the mystery of how we had a perception that could be analyzed in so-and-so way from the start. To neglect this fact is to forget the “non-knowing of the beginning which is not nothing” (Merleau-Ponty 1968, 49). By neglecting the perception prior to reflection which founds that reflection, we waive off the perception as “literally nothing” (this might end up just being a rephrasing of the final pages of the introduction to Phenomenology of Perception):

… every resistance to its [reflection’s] exercise being from the first treated not as an adversity of the things but as a simple state of non-thought… about which there is nothing to say since it is literally nothing. (Merleau-Ponty 1968, 44)

Additionally, they are thoughts about things, not the things themselves. I think the point is that our reflections render the world into an idea, or a thought, whereas the world cannot be reduced the subject’s thoughts. In other words, a philosophy of reflection believes that the world and thoughts about the world coincide (e.g. “the being-self is given to me in the adequation of my thought with itself” (Merleau-Ponty 1968, 48)) (whereas they do not). (This is why, I think, MP regards philosophy hitherto as playing with word-meanings: to signify is to essentialize (to fix particulars under a universal), and reflection is all about essentializing.) This is a holdover from or transposition of perception, an insight we see from the Phenomenology: as appearances make way for further appearances, we take what we see now as what has always been despite previous appearances being phenomenally different. Analogously, we take our analyses about a thing to be identical with the thing analyzed. In other words, philosophies of reflection think they can retrace the steps back to an original unreflected perception (which I think the Phenomenology assumes!, which is why I take this to be a critique of the mid-MP), but such a philosophy is circular because what it finds at the origin is what was put into it by this reflective analysis (I think).

The above two halves of every philosophy of reflection can be consolidating into this remark: philosophy of reflection is of “an order of idealization and of the ‘after-the-fact’ which is not that wherein the world is formed” (Merleau-Ponty 1968, 45). Both points are two sides of the same coin: there is a characteristic distance between the world and our reflection (thoughts about perception). (Consequently, the subject and the world are merely “pure correlation” (Merleau-Ponty 1968, 47), that is, merely parallel regions; in actuality, in a Hegelian move, they are internally dependent as opposed to externally related.) Indeed, reflection requires that one take a step away to look up the unreflected thing, which is to say, that the subject must detach itself from their participation in the world to reflect upon it:

… every effort to comprehend the spectacle of the world… demands that we detach ourselves from the effective unfolded of our perceptions and from our perception of the world, that we cease being one with the concrete flux of our life in order to retrace the total bearing and principle articulations of the world upon which it opens. (Merleau-Ponty 1968, 45)

Hence why Husserl realized that “every transcendental reduction is also an eidetic reduction” (Merleau-Ponty 1968, 45).

So we need to recognize the pitfalls of reflection and overcome it, finding the “secret of our perceptual bond with it [the world]” (Merleau-Ponty 1968, 38).^[2] How do we do this? MP doesn’t resort to a naive rejection of reflection; instead, he thinks that we should take philosophy of reflection as one stage that is useful but has its flaws and should be advanced from. (This makes sense, since he uses reflection to discover the pitfalls of reflection, as far as I can tell.) This is where I believe the influence of Hegel is strongest, and where I think he most clearly demonstrates an evolution from his analysis in Phenomenology of Perception. It seems MP’s point is that there is perception, perception-reflected-upon, the thing, and the thing-perceived-within-perception-reflected-upon, and instead of trying to recover perception (the unreflected) and the thing(-in-itself, as opposed to the phenomenal thing), we should create a philosophy of hyper-reflection (elsewhere he describes this as a “philosophy of total reflection” (Merleau-Ponty 1968, 43) because it takes into account the total situation, that is, perception and reflection-on-perception conjointly): reflect upon how a perception is simultaneously accompanied by a reflection about that perception which is necessarily at a distance from it yet coexists with it. A philosophy of total reflection, then, will be able to contact the two problems a philosophy of reflection leaves untouched: understanding the “effective world” and how we can ideate about the world via reflection (Merleau-Ponty 1968, 43).

Mazis, Glen A. 2022. “Merleau-Ponty’s Embodied Ontology and Literature: Gesture, Metaphor, Flesh, and Sensible Ideas.” Humanities 11 (1): 26. https://doi.org/10.3390/h11010026.

Merleau-Ponty, Maurice. 1968. The Visible and the Invisible. Edited and translated by Claude Lefort. Edited by Claude Lefort. Translated by Alphonso Lingis. Studies in Phenomenology and Existental Philosophy. Northwestern Univ. Press.