Red, Mary and the Appreciative Act

Mary’s Room, Duration and the Layering of Wholes

An opening

In his second lecture Iqbal observes that in a single glance at something red we hold together a wave frequency that would take thousands of years to count, and that this is how "the mental act transforms succession into duration” (Reconstruction, p. 39). The remark is brief, but it contains a whole theory of experience. This essay follows one question that grows out of it. If red can be counted on one side and felt on the other, what exactly does a person gain the first time they see it? From that question the essay moves to machines, to chemistry, to evolution and, last, to the small daily panic of being late.

I proceed as a conversation proceeds, one worry leading to the next. Where I borrow, I say so. Where I am not sure, I say that too.

Mary and the missing bridge

Frank Jackson’s thought experiment imagines Mary, a scientist who knows every physical fact about colour vision (wavelengths, cones, the pathways of the brain) but has lived her whole life in a black-and-white room. When she steps out and sees a ripe tomato, does she learn something new? Jackson says she does. David Lewis answers that she gains only abilities: to recognize and to imagine red, as one gains the knack of riding a bicycle.

Consider what recognition requires. Place two coloured patches before Mary. She can tell at once that they differ, since telling apart needs no name. But to say "this one is red and that one is blue” she needs a bridge between a described thing and a lived one, and her physical knowledge cannot supply it. Someone must tell her, or she must test it herself. A well-equipped laboratory seems to remove the difficulty: she shines light of a known wavelength, looks, and writes down "this is what I shall call red.” Yet notice where the bridge came from. It came from looking. The laboratory did not replace the experience. It arranged for the experience to happen under controlled conditions.

The moment of matching, the small eureka of "so that is what it looks like,” is not a trick of the imagination. Something has been learned that the theory could not hand over. The ability hypothesis says only that Mary’s new knowledge is a skill. But the skill of recognizing red as red cannot be built from the theory alone, which is exactly what is in question.

Counted and felt

Bergson gives this gap a name. In Time and Free Will he separates quantity, which is laid out in space and can be divided and compared, from quality, which is lived in duration and cannot be divided without being destroyed. In Matter and Memory he adds the example Iqbal later echoes: the sensation of red contracts a vast number of vibrations into one undivided moment of perception.

Seen this way, Mary’s knowledge lies wholly on the counted side. A wavelength is a number. Red as lived is a whole with no parts to number. The two are not the same fact described twice. They are two modes of knowing, which Bergson calls analysis and intuition, and Iqbal calls efficiency and appreciation. The efficient self sees the 700 nanometres, and the appreciative self sees the red.

One caution belongs here. To say that colour is duration explains why red cannot be analysed into parts. It does not explain why that particular rhythm feels like red and not like blue. Bergson would reply that the question wrongly asks for a part-by-part derivation of something that has no parts, but critics find the answer evasive, and I think they are partly right. The idea clarifies the problem without solving it.

Compression is not contraction

A modern system can transcribe an hour of speech in seconds. Is that a contraction in Bergson’s sense? I think not, and the reasons are instructive.

First, a recording is time already turned into space. The hour exists as a row of samples, all present at once, and the machine reads the row. It does not wait through the hour, and Bergson’s glass of sugar water makes the point: the waiting is part of the phenomenon. Second, compression keeps its parts. Even lossy compression works by dropping or merging pieces, and the output is still made of pieces that can be compared and recombined. Contraction cannot be reversed in this way. You cannot recover the trillion vibrations from the red, and a remembered melody is not its notes played faster. Third, nothing is lived. Play the same file twice and the machine returns the same result. A person hearing the same hour twice hears something different the second time, because the first hearing is now part of them.

This shows that transcription is not duration. It does not settle whether any artificial system could have duration, and I would distrust a confident answer in either direction. The relevant test would be whether a system’s whole past is present in each moment, irreversibly, and not stored and retrieved. Speed is a property of the counted side and is no evidence either way.

Chemistry and stakes

Suppose one gave such a system the chemistry of feeling: neurotransmitters, hormone-like signals, slow diffusing messengers. Artificial synapses that respond to real neurotransmitters already exist in laboratories, and global neuromodulation signals are common in software networks. The engineering is hard but not mysterious.

Chemistry alone, however, is still describable as molecules and concentrations, which places it on the counted side. What may matter more is stakes. Antonio Damasio and Anil Seth argue, in different ways, that feeling is bound up with the regulation of a living body: hunger, temperature, damage, survival. Hormones matter in animals because they serve that regulation. A chemical signal that governs nothing the system must preserve is ornament. And even if every ingredient were present, we would have no means of checking whether anything is felt. We would see behaviour and measurement, which is Mary’s predicament again, now about the machine.

Two tempting shortcuts should be set aside. The delayed-choice quantum eraser does not show irreversible information, since its whole point is that which-path information can be erased until it leaks into the environment. And quantum computers are engineered to avoid exactly that leak until the final measurement. The claim that consciousness depends on quantum effects in the brain, advanced by Penrose and Hameroff, remains a minority view, and Tegmark’s estimates of decoherence times in warm neural tissue are a standing objection. Irreversibility itself needs no quantum mechanics. A dropped egg does not reassemble.

Wholes upon wholes

Iqbal’s appreciative act is a whole grasped at once. The ball is caught in one gesture, the colour seen in one glance, the melody heard in one phrase. My suggestion is that these wholes are layered. There is a whole for red, another for vibration, another for touch and smell, and over them a whole for the situation, and over that a whole for the self that is in it. Self-apprehension would then be the top of an iceberg whose lower parts reach down to the earliest reflexes.

Such layering is not foreign to current science. Damasio describes a proto-self, a core self and an autobiographical self, each resting on the one below. Hofstadter calls the self a strange loop. Higher-order theories of consciousness, such as Rosenthal’s, hold that a state becomes conscious when another state takes it as its object. My own formulation is that earlier appreciative acts become the objects of later ones.

That sentence also accounts for measurement. To measure, one must treat a whole as an object. A creature that can apprehend its own apprehending can step back from the flying ball and ask how far and how fast. The efficient act is then not the opposite of the appreciative act. It is the appreciative act turned on itself, which agrees with Bergson’s remark in Creative Evolution that intelligence is surrounded by a fringe of intuition.

The evolutionary story needs a light touch. Evolution has no ladder of higher and lower species, and humans do not have more senses than other animals in any simple way. Other creatures sense magnetic and electric fields that we cannot. What marks us is flexible tools, language and culture. These change the environment, which changes selection on the next generation, a feedback sometimes called gene-culture coevolution. Self-recognition also appears to have grown by degrees, since chimpanzees, dolphins, elephants and magpies pass versions of the mirror test. Whether self-apprehension was genetic, epigenetic or something else is therefore best answered with all three, together with culture.

A final warning about metaphor. A Fourier series builds a wave from separate sinusoids that add up, and that is analysis, the counted side. Layered wholes are not sums. Each layer reorganizes the one beneath it, and nesting is a truer picture than adding.

Does the earlier act survive underneath intact, or is it altered by being taken as an object? Bergson would say it survives, and that intuition is the way back to it. If it is altered, the structure of the iceberg is different. I leave this open, and I suspect it is the question on which the whole account depends.

Why we hurry

The same structure explains a common unease. When I am late for the office I brush my teeth without attending to it, because my attention is already at the next thing. The act proceeds by habit while the self is elsewhere.

Research on attention supports the description. The mind holds a representation of the next goal while performing the present task, and when the goal feels urgent it takes the attention. Unfinished tasks stay active in memory more than finished ones, a result associated with Bluma Zeigarnik. Time pressure raises stress hormones, which narrow attention onto the threat. In cardiology the habit has been called hurry sickness, and in the psychology of flow its opposite is full absorption in the present act.

In the terms of this essay, the plan has divided the morning into parts, and I begin to live in the list instead of in the morning. The efficient act overrides the appreciative one. The destination is only an imagined next step, and when I arrive the image has been replaced by the one after it, which gives the hurry something of the quality of waiting for Godot. The toothbrushing still takes its own duration. Rushing sacrifices its quality, not its length. Some hurry is real, because the train does leave, but the extra hollow part is attention that left the present when leaving would not have helped.

A closing

Three claims have been made, with differing confidence. First, that Mary learns something when she sees red, namely how the described thing appears from within, and that no laboratory removes the need for the experience. Second, that this gap is better understood as a difference between two modes of knowing than as a missing fact, a view Bergson and Iqbal share. Third, that appreciative acts are layered, that each layer becomes the object of the next, and that measuring, tool-making and self-awareness are consequences of this turning back. The first claim is the most secure. The third is the most speculative, and the one I most want to test.

References

Primary

• Bergson, Henri. Time and Free Will (1889). Trans. F. L. Pogson. London: Allen & Unwin, 1910.
• Bergson, Henri. Matter and Memory (1896). Trans. N. M. Paul and W. S. Palmer. London: Allen & Unwin, 1911.
• Bergson, Henri. Creative Evolution (1907). Trans. Arthur Mitchell. New York: Henry Holt, 1911.
• Iqbal, Muhammad. The Reconstruction of Religious Thought in Islam. Ed. M. Saeed Sheikh. Lahore: Institute of Islamic Culture, 2011 printing.

Secondary

• Damasio, Antonio. The Feeling of What Happens. New York: Harcourt, 1999.
• Hofstadter, Douglas. I Am a Strange Loop. New York: Basic Books, 2007.
• Jackson, Frank. "Epiphenomenal Qualia.” Philosophical Quarterly 32 (1982): 127-136.
• Lewis, David. "What Experience Teaches” (1988). In Papers in Metaphysics and Epistemology. Cambridge: Cambridge University Press, 1999.
• Rosenthal, David. "Two Concepts of Consciousness.” Philosophical Studies 49 (1986): 329-359.
• Seth, Anil. Being You: A New Science of Consciousness. London: Faber & Faber, 2021.
• Tegmark, Max. "Importance of Quantum Decoherence in Brain Processes.” Physical Review E 61 (2000): 4194-4206.