A description is a way of dividing the world into at least two parts: the phenomenon to be described, and the phenomenon used to do the describing. If we describe a rose as red, we implicitly divide the world into the rose on one side and everything else (such as concept of redness) — on the other. For a description to be useful, it must convey information that the listener doesn’t already possess. Saying “this rose is rosy”, is redundant. We can label the two parts of a description the target (the phenomenon to be described) and the descriptor set (the set of concepts used for describing).
Imagining the world as split into two parts helps us recognize the structure of description, but to further explore how scientific description works, we will benefit from splitting the world into three. This third part is the objective (or intersubjective) observer. A scientific description of a rose involves parceling the world into the rose, the concepts used to describe it, and the observer who is the recipient of the description. Even when a solitary scientist in the wilderness is describing something to themselves, this idealized observer is present and listening.
The virtual presence of the observer forces the scientist to describe things from a third person perspective. They must say things like “the rose is red”, rather than “the rose is like the one in my garden” or “I see the rose”. Even when a scientific description involves first person statements, it must be translatable into a third person view. The statement “I measured the wavelength of the light bouncing off the rose, and it turned out to be 680 nanometers” is in the grammatical first person, but the underlying assumption is that anyone who measures the same rose under the same lighting conditions — conditions that must also be described — will get the same result from the act of measurement.
A third person description can sometimes obscure the presence of irreducibly first person subjectivity. One might, for example, say that the rose is beautiful. But how does one describe beauty? The vagaries of taste render this description hard to objectivize, as generations of philosophers have learned. Even ostensibly physical descriptions may prove to be subjective — or at the very least under-specified. Imagine someone describing the rose as “large”. Perhaps it is the largest rose the person has ever seen. Other people do not know what the experiential history of the describer is. The concept of largeness can only become objective if it is ‘calibrated’ in relation to an object that the idealized observed has access to. (This is obvious to many of us now, but it wasn’t always so: understanding that movement is not an intrinsic property of an object, but a relation between an observer and the object, was a crucial breakthrough made by Galileo, and essential to the scientific revolution.)
This ‘conceptual calibration’ has always been the basis for measurement — we compare a thing of interest with some kind of standard. It is worth noting that the standards by which we describe an object of interest cannot be simultaneously described along with the object, unless we use additional standards that are not currently subject to description.
Most people are aware that the target of description should not be part of the descriptor set — saying “the rose is red because it is a rose” is not a helpful description. Smuggling the target into the descriptor set is analogous to circular reasoning — it assumes familiarity with the very concepts being described.
A scientific model is a descriptor set: it is a segment of the world used to describe a distinct segment of the world to an idealized observer. It cannot be used to explain itself, because that would be circular. A model is like a map: distinct from the territory that it represents. And for pragmatic reasons it is simpler than the territory; any map that is a perfect representation of a region would be coterminous with it. Such a map is completely useless (and “not without some Pitilessness”). Scientific models, like maps, are designed to be simpler than the things they describe.
In certain strands of Buddhist discourse one encounters the line, “don’t confuse the finger pointing at the moon with the moon itself.” It resonates with two other sayings that are common in scientific circles: “the map is not the territory” and “all models are wrong but some are useful”. But the finger-moon advice might be more effective at conveying some key ideas.
Scientific models acquire tangible form when expressed in words, mathematical symbols and diagrams. Only rarely does the physical form of a model resemble the phenomenon it describes as closely as a map resembles the territory it represents. Nevertheless, many people would consider the relation between model and phenomenon to be roughly the same as the relation between map and territory: a structural analogy. In sharp contrast, the finger and the moon are not structurally analogous at all. What then might we gain by thinking of a scientific model as a finger pointing at a phenomenon? Instead of similarity, the operative concept becomes pointing.
Even when we assert that a map is not the territory, or that the model is not the phenomenon, the very nature of mapping (or perhaps of the copula “is”?) invites us to wonder about a perfect isomorphism between model and phenomenon. If we had a theory of everything that was isomorphic with reality, we would be justified in claiming to know what things are, rather than what they are like, or where they are in relation to other things. But in the case of the finger and the moon, we do not expect the act of pointing to imitate reality in the same way. Pointing is a gesture that draws a person’s attention to something. When it is successful, the gesture fades away and the intended target dominates the observer’s attention. Pointing is a tool for grasping some aspect of reality, rather than for creating a cognitive photocopy.
Can we accept the idea of science as a tool for attention, rather than imitation? Admittedly, this is a radical departure from the standard picture (!) of science. No one would assert that science is just a tool for grasping; anatomical diagrams, for example, have a clear resemblance to the things they represent. The suggestion that science is a finger pointing at the moon must therefore be taken as a provocative speech act — it draws our attention away from the nimble hand that is so good at imitating other things, and towards the act of pointing itself, which in a scientific context means the act of aligning elements of a model with elements of a phenomenon.
The tripartite division necessitated by science — which splits the world into target, descriptor set, and idealized observer — might render certain “big” questions paradoxical or meaningless. For example, if this division is really necessary, then is a theory of everything really possible? Once everything is the target, what remains in the descriptor set for us to describe it with? Can we even describe the act of description, which relies on the idealized observer? Is it possible to partition reality in such a way that the observer is both left out (so as to be the recipient of description) and left in (so as to be the target of a description)?
In other words, can the finger pointing at the moon also point at itself?
Note
I initially thought of using the terms “explanandum” and “explanans” for “target” and “descriptor set”. But the word explanandum is typically used to means the description of the phenomenon, and is distinguished from the phenomenon itself. The explanandum is, in other words, already the finger pointing at the moon! See how tricky this can be?
Further reading
For a related perspective on the nature of science, including the status of the observer, see this essay I wrote for 3 Quarks Daily, which is structured around the concept of symmetry/invariance:
Science: the Quest for Symmetry
Implicit in the act of description is the process of naming. My very first 3QD essay explored this from the perspectives of Taoism and cognitive science:
Boundaries and Subtleties: the Mysterious Power of Naming in Human Cognition
Charles Sanders Pierce’s concept of thirdness might be related to the tripartite division I discuss here (but I’m not sure yet, so correct me if I’m missing something). This essay by a philosopher named Brian Kemple discusses the concept:
The Continuity of Being: C.S. Peirce’s Philosophy of Synechism
If every rose was an individual rose there would be no rose and each of them would have another name. But then there would be no naming of anything. The finger points at the moon but can we point at pointing. As in the childrens’ card game of snap, images flow over each other and await congruence.Well, you may say, pointing, mapping, comparing are meta activities supervenient on the data. True but can they be taught?
A response: It is not necessary to know what we are doing. ‘Doing is the mother of doing’ said Samuel Johnson. So that is what I was at. Comparison is one of the Pramanas or a valid means of knowledge that is underived. This is a scandal for the Empiricist.
I completely agree! Comparison is simply named, rather than described. But we can point to different types of comparison, once the basic idea is established.
The idea of the individual rose has been bothering me lately. I think I was an unthinking nominalist until recently — assuming only particulars “exist”, rather than categories. But lately, upon reflection I’ve been finding this somewhat incoherent. In a universe of spatio-temporal particulars, nothing can be named — and perhaps nothing can be perceived either, given the importance of top-down processing even in low-level perception. I suspect the universal and the particular are bolted together in dialectical fashion (perhaps this an example of pratityasamutpada?).
This might also relate to the difference between “seeing” and “seeing as”. The latter is woven into our phenomenological field, and it very hard to suppress. Realism lurks in Pierce’s ‘thirdness’.
Recently I was intrigued by the realism of Kumarila Bhatta, who opposed it to the nominalism of Buddhists (and I think Advaita Vedantists too). I get a kick out of the idea of “real absences”. I discovered this debate in an episode the History of Philosophy in India podcast (hosted by Peter Adamson and written with Jonardon Ganeri), so I haven’t yet read anything books or papers about this.
I really like this image, by the way:
“As in the childrens’ card game of snap, images flow over each other and await congruence.”
Tricky indeed. I enjoyed the recursion of the paragraph that starts, “Can we accept the idea of science as a tool for *attention*, rather than imitation?” and continues, “The suggestion that science is a finger pointing at the moon…draws our attention away from the nimble hand that is so good at imitating other things, and towards the act of pointing itself”. In other words, the suggestion itself is a form of scientific description in your model (which is itself a scientific description of a scientific description of a…). Which is to say, all scientific description (to the extent that it maps to your description of it) is a provocative speech act!
I’m so happy you noticed that recursion! Makes writing really worthwhile when people notice ideas like that, which are sort of hidden between the lines. 🙂
Isn’t it the case that scientific measurement and observation is largely the enhancement of normal perception? Microscope, telescope, balances, deflections on a gauge etc. Certain events seen by the naked eye can open up an avenue of esoteric mathematics. I’m thinking of this phenomenon. re (rotating bodies, wing nut behaviour in zero gravity)
https://www.youtube.com/watch?v=1VPfZ_XzisU&t=289s
My point is that scientific realism/primary qualities (Locke) are not so divorced from the secondary subjective qualities. The idea that this table is mostly empty space is sophistry to awe the lay. Empty space at table scale is where you might put bookshelves.
Haha I saw that video recently too. Anyway, if I understand you correctly, then yes, science is an extension of sense.
“The idea that this table is mostly empty space is sophistry to awe the lay.”
Yup. It’s an example of people taking precisely the wrong lesson from science. Another example is saying that people who think the earth is at the center of the solar system are stupid. You can transform coordinates however you feel like. Plus the sun also moves, so if you are really nitpicky you can claim that from a perspective not fixed to the sun, the solar system moves in a kind of helix through space. The real lesson of Copernicus is that the number of free parameters is reduced compared to the epicycle perspective.
But the notion of simplicity in fitting data is far too instrumentalist for the missionaries of scientism — they want ontology too!
I really like the idea you are putting across of somehow using the finger in the metaphor of ‘the finger pointing at the moon’ as our scientific model(s) of reality and explore what this could mean. It got me excited about where this could be pointing at (pun intended).
But first let me tell you about a personal story I have with this metaphor.
Long before I got married, I knew intellectually ‘the finger pointing at the moon’ metaphor. About thirty years ago, I was playing a simple game of rolling a ball back and forth with my one-year old daughter in our sitting room. At one point the ball rolled and went behind the sofa chair, instead of directly towards her. She looked to her left and right searching for the ball but could not see it. I got up, walked, and sat beside her. Then I pointed to where the ball was, but my daughter did not look to where I was pointing and looked at my finger instead! No matter how much I tried to point to the ball she could only look at my pointing finger. Then I realized that my daughter was too young to understand the meaning of pointing. And then I also realized that my intellectual understanding of ‘the finger pointing at the moon’ was still immature to understand the meaning of pointing in the spiritual realm.
The metaphor is made up of the moon, the finger, the act of pointing and the observer to whom the pointing is used to direct his looking or attention. But the metaphor is a process consisting of (one being unaware of something) + (one sees the pointing finger) + (one follows with his eyes along the line where the finger is pointing) + (thus abandoning the finger and seeing or becoming aware of the thing being pointed at). It is like the bridge that takes you from one side of the river to the other side which is similar the Buddha’s parable of the raft where one needs eventually to abandon the raft after crossing the river to the other side.
Thus, “don’t look at the finger pointing at the moon, look at the moon” enjoins one to consider the finger as a mere pointer which needs to be abandoned by the observer in order to look at the moon.
For the scientific model to be like the pointing finger one needs to use the model like a bridge or a raft to cross to the other side of reality and then abandon the model.
One possible interpretation of this is to not be caught up by the ability of science to totally understand or perceive the whole of reality.
The metaphor of ‘the finger pointing at the moon’ might prove to be detrimental to the body of science as it is practiced now because it implies that one must eventually abandon present day science for something more true, or a science that is more systemic and holistic; perhaps, something like the ideas in the book “The Systems View of Life: A Unifying Vision” by Fritjof Capra, but not for just living systems but for everything in the universe.
I am wondering; Are you trying to say that we need to use our scientific models of reality as pointing to some greater or deeper reality in order to grasp which we need to abandon the model(s)?
Very interesting! My primary goal was to show that pointing does not require similarity between the index (finger/model) and the thing (moon/phenomenon). As I said, scientific models often do contain “representations”. But I guess you could say that technology represents a kind of “abandoning” of the model — a model that works very well disappears from cognition, and is buried in the technologies that offer grasp over some little corner of nature.
Any mystical point can only be made indirectly, without pointing directly at it. 😉