This is a decent analogy, to an extent.
Lmfao.
Not entirely true. The evolutionary perspective is only one approach to understanding the brain, and is something that is actually very well understood at this stage. This is true for the functionalist approach as well (think computational theory of mind).
____________________________________________________________________
RE: OP: We actually have a ton of knowledge about the brain so I don't know what you are talking about. A "serious" mapping of a brain is both feasible and extant. The only issue is that it is a mapping of a prototypical brain - one that is generated as an example of a normal brain, not an actual individual's brain.
It all depends on what kind of map you want. If you're aiming at what I think you are, which is neuronal network mapping, then the issue with mapping a real human's brain is that there currently isn't an entirely safe way to do so that isn't also incredibly painstakingly slow and reliant on patient feedback. It also depends on how much detail you want to go into, as well as what kind of map you really want - just a bunch of roads? Or do you want the roads and destinations to be named? If so, why do you think that is feasible given the absurd amount of neurons that we have, and the fact that while the average person may have similar "areas" of the brain with similar functions, their neuronal configuration and specialization is unique to them?
Let me put it in simpler terms.
Think of a 10 x 10 grid:
I know this isn't 10x10 but this is just a visual aid.
There are 100 squares in a 10 x 10 grid.
I know this isn't 10x10x10 either.
Now, if we make that grid into a box by putting a grid vertically next to it and expanding them both, you will get 1000 cubes. Let's imagine that every 3D corner of a little cube is a neuron; keep in mind that there are 8 corners to a cube, even though all or most are going to be shared with their neighboring cubes. That is a lot of corners!
Let's do it in a more direct way that involves less complicated math.
Manhattan, New York City, and all of its streets. Now make picture a box, with each one of faces of the box being a map of Manhattan. Now, fill the inside of the cube by projecting every street that is on the surface through to the other end of the cube. Now, at every point where there is an intersection, we have to name that intersection.
On Manhattan island, there are
11,485 street corners, with about
700 streets making them. That means that if we assume that each side of Manhattan has the same amount of streets (it doesn't but bear with me), then that means there are about 350 streets on each pair of sides (because a street runs from one side to another). In order to have a decent guesstimate as to how many 3D intersections our box would have inside of it, we need to multiply the 2D figure (11,485 corners) by the "1D" figure of a single side (350). This yields 4019750 total 3D intersections - a model of over 4 million neurons. Imagine having to map every single one, and label what it does and what it's responsible for and what the various pathways are! For example, the neuron at (10,122,34) on the (x,y,z) plane has meaningful connections to neurons at both (56,13,270) and (234,120,16), each path being responsible for a different thought process. How are you going to "map" each one of those connections in a reasonable way?
Now, keep in mind - this is only 4 million neurons. Do you know how many neurons we have? We have
25,000 times more than that, for a total of
100 billion neurons.
So, this begs the question, if indeed this is the kind of mapping you think the government has hidden from us: is it because YOU know so little about the brain that you assume that all of humanity does, and thus why you think that a map of over 100 billion individual neurons in each individual human being is feasible with the technology that is available today?