Free-market solutions to saving the environment?

December 28, 2011

The new Republican flavor of the week is Ron Paul, and since he looks like a contender to win the Iowa caucuses, I’ve been reading up on libertarianism and its proposed solutions on several issues that critics historically identify as failings of the free market.

I think that Ron Paul has not properly addressed the issue of climate change and environmental protection. He claims to have approached the problem “the same way [he] look[s] at all other serious issues: as objectively and open-minded as possible.” His interpretation of climate data, although a step up from his fellow party members, is still wishy-washy. Maybe, considering the results of the recent Berkeley Earth Surface Temperature Study might warrant a second look from Mr. Paul.

However, many libertarian solutions to saving the environment are intriguing, and they might end up saving us. While most Western countries’ governments debate the best ways to combat climate change, Congress seems to spend its time arguing about whether it’s even real.

And what if it isn’t? I remember a quote from somewhere a few years ago, which went something like, “Maybe one day, when we’re independent of foreign energy and our environment is clean and well-preserved, it’ll turn out that global warming was a hoax after all. Then we’ll say, ‘Man, those liberals really got us, huh?'” There are a lot of free-market incentives to searching for sustainable and environmentally friendly energy sources. Solar power can be cheap, abundant, clean, and non-political. Much of the turmoil in the Middle East would effectively end, and corrupt and repressive dictatorships from Venezuela to Russia would collapse.

Solar power doesn’t even have to be a cute little thing that wealthy liberal Western European countries such as Germany or Denmark subsidize – according to this article, solar power could be, in terms of efficiency and cost, a better alternative to hydrocarbons in just a few years. Cost of solar energy has been decreasing exponentially, or approximately halving every two years (a similar trend as Moore’s law regarding the number of transistors on a computer chip). Meanwhile, oil isn’t getting any cheaper or tech-savvier.

This is still largely speculative. Free-market environmentalism doesn’t properly address many topics such as protecting endangered species (there may not be strong incentives for the majority of individuals to do so),  or negative externalities in general. A lot of environmental protection involves solutions that are realized over a long period of time, or don’t allow a do-over – two situations which historically have not been the free market’s strong points. However, maybe it’s not too naive to hope that the world may be saved by those who would profit immensely from doing so.

Poverty of The Stimulus and the Rationalist/Empiricist Debate

December 28, 2011

(Note: This is my second post in the series about the philosophy of cognitive science)

Imagine that you’re a traveler passing through a small village in a foreign land, and as you’re trying to communicate (with difficulty) with the locals, you see a rabbit running by you. As it runs by, a local says, “Gavagai!” Now, of course, you might infer that he was referring to the rabbit. So does “gavagai” mean rabbit? Could it mean furry? It could even mean Nice Day! You don’t know.

This dilemma was devised by Van Gordon Quine (1960) to refer to something called the “poverty of the stimulus” problem – under-determination. If we consider the brain to be a sort of information-processing device, an instantiation of the abstract concept of a “mind,” then researchers are dogged by the issue of language-learning, and all kinds of inductive learning in general. Noam Chomsky, renowned linguist and cognitive scientist, framed the problem thus – children can learn a grammar which can produce an infinite number of sentences by only hearing a finite number. How do children possess the universal and consistent ability to learn languages, using a relatively small amount of data to learn an internal representation of language capable of making sense of unfamiliar languages?

Chomsky believes that certain aspects of language must be innate. According to him, certain (universal) aspects of language are encoded into our genome. Important as this question is, it in turn raises bigger ones – to what extent is our knowledge and mind predetermined?

As Dwight Schrute from The Office might say, there are basically two schools of thought. The first is rationalism, the philosophy based on the belief that our mind comes equipped with innate ideas. The  second is empiricism, which believes our mind is fully shaped by experience.

Empiricism, influenced greatly by the thinker John Locke, promotes the idea of a tabula rasa, or blank slate – at birth, we are all equal, and our minds are fully shaped by the experiences and actions of our lifetimes. It’s an attractive idea, and one of the tenets of libertarianism, the political philosophy of which Locke was in many ways an intellectual ancestor. With empiricism, the entirety of knowledge is based in ideas. Some ideas are direct, others are abstract and indirect. Mathematical concepts, such as the triangle, would be considered “abstract.” Although some would argue that triangles, as well as other abstract concepts, might be impossible to properly represent in an idea-scheme, some empiricists such as John Stuart Mill went so far as to proclaim that all of mathematics COULD be depicted in terms of definitional relations between ideas. The purpose of reason and logic, then, were to organize ideas. One important empiricist belief is that the mind is a domain-general device, in that it uniformly picks up and learns from experiences and stimuli.

Rationalism is quite the opposite – according to this philosophy, the mind comes equipped with innate ideas. Essentially, the mind takes an axiomatic approach about learning about the world, with the axioms built in. Some knowledge is not derived from experience. This is because many ideas, such as identity, must be inherent, or otherwise we wouldn’t have any ideas at all. This is where Descartes’ “cogito ergo sum” comes in. This is also where Descartes’ dubious proofs of the existence of God come in – as I mentioned in a previous post, he believed that since we had an idea of God, it had to be implanted in us innately. And what force could possibly provide an idea of the existence of God? Why, God him and or herself. Also, the rationalists believe that the mind is a domain-specific device – because of innate characteristics, it is specially structured to learn from different stimuli differently.

So which is it, empiricism or rationalism? This is, at a base level, the philosophical equivalent of nature vs. nurture, and in many ways, not very practically applicable, because it doesn’t tell us much anything useful. The ideas of the innateness of language and the modularity of the mind (we use different parts of our brain to respond to different situations, eg. we’re really smart if we suspect someone might be cheating) are scientific, not philosophical theories. The rationalism/empiricism debate is an important episode in the history of the philosophy of the mind, but it went back to a time when thinkers didn’t make a connection between the actual physical structure of the brain and the structure of the “mind.” Now, from a scientific perspective, we cannot proclaim rather little, because nature vs. nurture is an ongoing debate with far-reaching societal, and even political, considerations (think libertarianism). Some scientific theories, such as the poverty of the stimulus problem, argue for a certain representation of the mind, but whether or not there are significant preexisting structures in inexperienced minds that produce variation or bias in learning abilities and predispositions is still an open question.

Stay tuned for my next post, on the problem of other minds and first- and third-person perspectives!


An incredibly simple proof of the halting problem (in Scheme)

December 20, 2011

Don’t know the programming language Scheme? Don’t have to! (Even if you *had* to it would take less than an hour to learn basic syntax).

When I began this blog I intended to make a blog post explaining a formal proof of Turing’s halting theorem (eg. no algorithm can exist which determines whether a computational procedure always halts for every input. For example, a program that took in a number and printed every number greater than that up to 100 would not halt for every number, for obvious reasons). Both Alan Turing, with Turing machines, and Alonzo Church, with the lambda calculus, independently proved the halting problem (“Entscheidungsproblem”). However, both of these proofs require a lot of background explanation, and they’re not the best way to do it anyways.

The best way, I found, was a simple exercise from the textbook, Structure and Interpretation of Computer Programs. Here is a construction of the proof:

Suppose we have “a procedure halts? that correctly determines whether p halts on a for any procedure p and object a.”

Then we write the following procedures:

(define (run-forever) (run-forever))

(This is an infinite loop), and then a specially constructed:

(define (try p)
  (if (halts? p p)

Now, imagine that we called  the expression (try try) and let it run. If (halts? try try) evaluates to true, signaling that the function try halts when called on itself, then try runs forever, a contradiction. Similarly, if halts? decides that try does NOT halt when called on itself, the function halts.

And there you have it – the contradiction! By reductio ad absurdum, a function such as halts? cannot exist.

This proof by construction can be implemented in any language where functions are first-class objects, such as Python, JavaScript, and many other modern ones (not Java, of course. Java kind of sucks).

And in addition, if you’re ever taking a computer science course which uses SICP as a textbook, Exercise 4.15 is answered for you!