See also website-ideas-science-puritySpectrumAndHandsDirtySpectrum .
Why things are difficult
Many parts of most fields of research are difficult because there is competitive pressure to go as far, as fast as possible, and it's mostly smart, motivated people working on this stuff, so at equilibrium all that would be left is stuff that is difficult to challenge these people. But there are different reasons why understanding something could be challenging, and so different fields of study have different sorts of 'bottlenecks' that make them hard.
- Philosophy: i feel like philosophy is often studying things that are hard to put into words. In fact, i think one of the functions of philosophy is to invent new language to allow us to talk about something that previously was very difficult to express. This is very difficult. First, because talking about something precisely using a language that isn't good at is difficult (duh). Second, because, since your initial formulation of the idea that is hard to express may be vague, it's hard to know if your idea that you want to figure out how to express is actually a real concept at all, or if it's just a bit of confusing fluff; and similarly, it's hard to know if something that another philosopher is telling you about is something real (that you just don't understand yet), or if it is just a bit of confusing fluff.
- Math, physics: these are hard because our brains, while capable of thinking in the necessary manner, aren't really 'tuned' for it. For example, when reading math or physics, i often find that i have to sit still, clear my mind, and focus intently on something i just read, going over it again and again and rehersing the meaning of the parts in my mind over and over until i can hold the whole thing in my mind at once and form an intuition about it. The slow rate of this process feels unnatural to me; most everyday things would be understood in a fraction of a second. I feel like my brain wants to 'time out', like a computer protocol abandoning a task after the server doesn't respond. The default timeout is too short for math.
- Neuroscience: this is hard because it's unclear what the ontology of the area is (what the 'natural kinds' are). That is, it's hard to know how many categories we should break stuff into.
- The problem is partially that many of the classifications that are useful in biology are subjective,
- and partially that there are just so many darn things that it takes humanity a long time to order all of them,
- and partially that there are different purposes for which it is better to group things into classes in different ways. Are we concerning ourselves only with how the brain computes, or also with its pharmacological responses?
- and partially that many of the experiments that we can do don't provide definitive results, but only results conditioned on some set of caveats, hence if we think that a particular class of thing has a certain property that may turn out to be incorrect if later work pulls the rug out under some of the assumptions that were made in the experiment that deduced that supposed fact -- and this may lead to a reevaluation not just of the properties of existing classes of things, but also of the way that we organized things into classes (for instance, if we decided that some earlier conclusions were wrong, this may lead us to reclassify the cells in some area into a different set of subtypes).
- and all this leads to enormous complexity because different authors use different ontologies and hence refer to the same thing using different language. Since the ontologies are different, however, one cannot simply say 'author A's term A == author B's term B'; it'll be more like 'the subset of author's A's term A where property C is true is a subset of author B's term B'.
- For example, we say that 'auditory cortex' is a certain part of the brains of many animals, and we classify cells as being 'part of auditory cortex' or 'not part of auditory cortex', but although a few candidate defintions of 'auditory cortex' may spring to mind, it isn't immediately obvious which, if any, of these should be used. Is auditory cortex all parts of cortex that respond to sound? Well, there are plenty of higher-order multimodal areas that 'respond to sound' in some sense, and on the borders of sensory areas some authors report multimodal cells, calling into question the notion of a discrete border. Is it all the parts that receive a certain type of projection from a certain part of thalamus? Is it all the parts which show a certain gene expression pattern? Is it all the parts that are descended from certain subpopulations of developmental progenitor cells? In fact, every individual's brain is different and the sizes and connectivity patterns of cortical areas differ, so is auditory cortex really an objective concept at all? Perhaps it is merely a repeating pattern across individuals which our mind tends to class as a distinct thing but that has no objective reality. This may seem extreme, although i don't think it is, but even if you do, you'll grant that a similar problem arises when trying to relate anatomical areas across species. How closely does the cerebral cortex correspond to avian pallium?
- Another example of this sort of thing is the question of how many celluar subtypes there are in a given area, and what they are. You'll see different authors referring to what you consider the same cellular subtypes using different terminology -- because the authors each have a slightly different classification scheme for these subtypes, so your opinion that there is an equivalence or simple mapping between them is only an opinion and is disputed.
- Another example of this sort of thing is the question of what the functions of the brain are. Vision and audition are clearly two separate things, but how about old long term memory and recent long term memory? Is the difference a matter of kind or of degree?
- As a result, much of the work in neuroscience to date is just figuring out the ontology of the objects of study -- e.g. what are the anatomical region, what are the types of cells, what are the types of functionality.
I imagine that philosophy might be the most difficult of all, because the difficulties i describe for it above seem the most insurmountable. But i haven't done much philosophy, so maybe i'm wrong.