Much of the thesis of Douglas Hofstadter’s I Am a Strange Loop revolves around the strange loop of self-referentiality that emerges when making “I”-assertions – these assertions reference themselves in order to describe what or who is being referenced. In other words: this “I” that we bestow so much meaning is all but empty – it is without content.
“We are powerfully driven to create a term that summarizes the presumed unity, internal coherence, and temporal stability of all the hopes and beliefs and desires that are found inside our own cranium – and that term, as we all learn early on, is “I”. And pretty soon this high abstraction behind the scenes comes to feel like the maximally real entity in the universe.”
According to Hofstadter, the strange loop of a self is an inevitable emergent consequence of any system that has a sufficiently sophisticated repertoire of categories – a hallucination hallucinated by a hallucination.
The first person subjective pronoun emerges as a convenient linguistic tool, but soon becomes imbued with meaning that grants the subject a sense of sovereignty in determining whoever or whatever they might want to be – a unitary causal agent full of desires – a Cartesian ego (I love this idea!)
In one of the chapters illustrating this illusion of private little islands of selves, Hofstadter references David Olesen’s 1964 wonderful pen-and-ink parquet deformation called “I at the Center”:
“Here one sees a metaphorical individual at the center, whose shape (the letter “I”) is a consequence of the shapes of all its neighbors. Their shapes, likewise, are consequences of the shapes of their neighbors, and so on.”
To me, this is a nice visual echo of the Buddhist notion of Pratītyasamutpāda (प्रतीत्यसमुत्पाद), or dependent arising.
Neither from itself nor from another,
Nor from both,
Nor without a cause
Does anything whatever, anywhere arise.
— Nāgārjuna's Mūlamadhyamakakārikā
The “I”s change shape the further away you go from the center in any direction, and the ways in which the shapes change also differ. Because this is a two-dimensional space, it’s not too difficult to see the variables that might inform the shape of a particular “I”: absolute distance from the center, and the relative direction from the center – the Cartesian coordinate system is enough for us to compute “I”-variations. For example, the “small vertical bounds” (the vertical lines on the floor and cap of the “I”s) are increasingly convex as you move left to right from the center, and tend to become increasingly concave as you move right to left from the center. Similarly, the “middle vertical bounds” are more concave as you move up from the center, and tend to become increasingly convex as you move bottom from the center, so much so that the lines even intersect.
Technical Notes
All the parquet deformations in this post, including the monochrome second one, a reasonably accurate representation of the pen-and-ink original, were generated using code I wrote. It’s written in JavaScript, using p5js, which is itself inspired by Processing.
The parquet is laid out by spacing out individual “I”-nodes, and deforming the shape of the various edges on each “I” proportional to the distance from the center: the peripheral “I”s are thus more skewed and exaggerated.
There’s definitely many other ways you could construct this parquet, some of which may include not even constructing individual “I”-nodes (which would be an even better metaphor).
Because my deformation is generated through code, you can also tweak the size of the “I”s, and the “skew factor” and generate all kinds of variations. Here’s a variation, with a smaller “I” and larger canvas size.
And here’s a parquet generated by randomizing the color of each “I” using the entire color space. The first image in this post was generated by restricting the color space. It’s fun to play around with variations to the original parquet, but I think the simplicity of the original is far more illustrative of the metaphorical individual.