Monday, October 31, 2011

Analogy paradox

I have come across many analogies since my higher classes. And I often wonder how easily people get entrapped in them even if the analogy is based on false assumptions, very close to 'Reductio ad Hitlerum'.

People use arguments to prove their point which are not really arguments but are kind of provocations. The analogy thus provided seems to be connected to the original point but in fact they have no connection. Examples of it can be as follows:

(Click on image for enlarged and clear image)
1) Quadrilateral analogy: In OOP (Object oriented programming), we have something called inheritance. Something that exhibit properties of something else need not be defined explicitly as it may inherit it from the other object. Like the properties of a lion is that of a Felis and all animals belonging to Felis are also mammals. Now if once the properties of Felis is defined to "mammals that are cat like", then all animals coming under the category of Felis will exhibiting the features of "mammals that are cat like" be it a panther or tiger or lion. The features of mammals are again to be defined explicitly as "Chordates who breast feed their offsprings". So, obviously, while stating the definition of Felis (mammals that are cat like), it is very clear that all felis are also mammals and they are "chordates who breast feed their offsprings". It need not be defined again. Also, primates may be defined as "mammals that are monkey like". And Gorilla then to be defined as a particular type of Primate means that it is also a mammal besides being a primate. So, all the features of Primate and a Felis in common are that of a mammal. The properties defined in 'mammals' are applied to both Gorilla ans Lion automatically when they inherit the properties from higher classes (here higher means more general i.e. Mammals is a general class, Vertebrates is a more specific and higher class, Chordates is even higher class and being an animal or even being a living being is even general term making it a higher class).

Same way, we have Polygons. The specific ones have certain features but also showing the features of more general ones, of which they are a part of.

a) Suppose we have a class in Java with signature 'geometricfigures'. Now I will define many kinds of objects under it like an ellipse1, ellipse2, circle1 etc.

b) Now I define another class 'polygons'. It will inherit all the features of a geometricfigures. Its objects may be like triangle1 etc. (Note that I could also have defined some figures like triangle2, square5 etc, but in that way, I could not say that triangle2  or square5 is a polygon. So it's better define them here. Still, I will ommit Square5 in this step as I know that I am going to make more specif classes).

c) Define a class 'quadrilateral'. Inherit properties from Polygons. Make specifications that Quadrilaterals have four sides. (This does not mean that quadrilateral has four vertices. As there are certain quadrilaterals which have four sides but 5 vertices (remember convex and concave?). You need to specifically define that It also has 4 vertices. Now the definition of our class 'Quadrilateral' becomes 'a polygon with 4 sides and 4 vertices'. Remember this definition is different than the real mathematical definition. As it is our program, we have our own definition of Quadrilateral. No problem. (But from now on we will stick to the mathematical definitions. I just wanted to omit the cases of confusions arising due to Covex and Concave Quadrilaterals etc).

d) Define a class 'parallelogram'. Inherit its some features from quadrilateral. Now specify other properties like 'opposite sides are parallel'. (This does not need to specify that opposite sides are also equal. That property is obvious, as no quadrilateral can exist where opposite sites are parallel but not equal. So, this property is an unwanted property which has stuck up even though we did not define it.)

e) Define class 'rectangle'. Definition will be 'a parallelogram' which has all angles equal to π/2. So, here we are with rectangles, also with features of polygon although we did not define it to have properties. Also note that properties of a rectangle like 'diagonals bisect each other' are not defined. They are stuck up with rectangle automatically.

f) Define class 'kite' under Quadrilateral (being 'under' means inheriting property from quadrilaterals). Specification other than being a quadrilateral will be 2 distinct pairs of adjacent sides being equal. So, diagonals bisecting at angle π/2 or angles between unequal sides being equal are obvious properties.

g) Define class 'rhombus' under kite as 'a kite with all sides equal.

h) Also define certain bojects like 'trapezium' and 'irregulars' under 'quadrilateral'.

Till now all we have is multilevel inheritance like
'geometricfigure' --> 'quadrilateral' --> 'parallelogram' --> 'rectangle'
and
'geometricfigure' --> quadrilateral' --> 'kite' --> 'rhombus'

i) Now I am going to define a class 'square' with no specific properties. All I am going to do is make it inherit its properties from rectangle and rhombus. So the definition of square will be 'a rectangle which is a rhombus' or 'a rhombus which is a rectangle'. This is an example of a multiple inheritance where One single class in inheriting properties from more than one class. Hence statements like this are meaningless (this reference is on Oct 31, 2011 if someone edits this definition in future for this Wikipedia article).

Click on image to view it clearly.
In the image above, the Red rectangles represent classes ehereas black, blue, violet and brown are objects. The light blue lines represent a flow of inheritances and grey lines represent another inheritance. With each advancing level, we define new properties with previous being intact. But in square, it was sufficient to just inherit properties from rectangle and rhombus as stated earlier.

Now comes the concept of analogy. Suppose, I provide a statement 'As diagonals of a kite bisect each other perpendicularly, the diagonals of a rectangle should also bisect each other perpendicularly because both are quadrilaterals afterall' does it sounds like a correct statement? Someone without the knowledge of figures and clear concept would fall for such an astonishing and lovely statement with lots of tech-jargon. But it is a pseudo-science. Though it seems to have a logical connection, in actual, there is none. Even if I give an example of an object, say square1 is a kite (as its adjacent sides are equal), its diagonals bisect at 90 degrees, now I also have proof to support my statement, but in actual, I will be neglecting the basics of classification.

Such an analogy is based on classic fallacy. In my view, the analogy is valid only if and only if (iff)
(i) the base of the two analogous facts is common (like being a quadrilateral is common in kite and rectangle)
(ii) and the analogy is limited to that common fact only (like limited to the properties of quadrilateral only)
no matter how many evidences you provided in support.

In the statement above, we followed the first rule of analogy perfectly, but we neglected the second one. And it made the statement an unreliable one. Even though it is possible that the given statement is true in all senses, but it might be due to some other facts, instead of the given analogy, as for an analogy to be valid, it has to be reliable in general sense, not in specific examples. Being true 'in particular' does not guarantee of being true 'in general' until and unless we do not deduce it mathematically.

2) Life after death and Evolution: It is also a classic example of 'Reductio ad Hitlerum' (though never heard from anyone else, its my view).  This argument is very common given by creationists against the theory of evolution. The truth is, that there is no link between the life after death experiences and the theory of evolution. Claim is that life after death proves the existence of god which further proves that 'if god exists, evolution can't exist'. Well this is not a meaningful statement at all as evolution has nothing to do with god. No where it points out that god does not exist or something like that. In fact, it is a very simple theory, easy to understand, but made too complex by creationists as they believe that accepting this theory will affect their religious faiths. If such people choose to live on faith rather than truth, that's their will. It doesn't change the fact that evolution is true and it has nothing to do with life after death experiences. Violation of my first rule. So invalidated analogy (in fact it is not an analogy at all, I believe).

3) Origin of life and big bang: A common question arises that where did we come from? That question is answered by biologists as 'your parents'. Where from they came? Well! their parents... and so on, back to millions of years, we reach the point that initially there must have been some some initial human. Where from did he come? Obviously, if you have not studied science (or if you are a kindergarten student), your answer will be 'Umm, someone must have made a human first. We don't know how he (or she) made him (her), or when did she (or he) made her (him) or anything else. So let's just assume his/her name is A and her/his partner be B (not respectively). Let the creator be C. Though, it doesn't look like fancy, lets rename A as Adam, B as Eve and C as God. Ahh!! That's nice.

But on reliance with evolutionary theory, it can be easily convinced that there was no first human as becoming a human was not an event. It was series of development. Something like asking 'who was the first engineer in the world' is an illogical question as being an engineer depends on the definition of what you call and engineer. Be it the inventors of plane, right brothers or before him Leonardo da Vinci or Galileo or aristotle. So, being human is not a specific question which include a set of properties and Voila, we get a human.

Now back to point. So, once you determine that humans came from evolutionary process, question comes in "where from did they come", and answer is even earlier animals. And so the question reverts back "where from did the first living being come?". Now the question has converted to 'What is the origin of life?' and it is no more the question of 'Evolutionary theory'. You have gradually shifted the focus from 'first human' to 'origin of life'. Still, the answer is there. Now, the question comes 'Ok, living beings came from so and so process from primordial soup. But where from did that soup come from?' And the answer goes back and back and finally reaches the Big bang.
Question: "Where from Big bang happen"
Answer: "We don't know"
Questioner: "Gotcha! You don't know what caused big bang. So, first humans were Adam and Eve. Hence proved"

Isn't that nonsense that if one can't prove Big-bang, you will deduce the first human to be Adam and Eve? The truth is, that there is no connection between the Theory of Evolution and the Big-Bang theory. But as the rejection of both of these theories support the religious scriptures written just few thousands years old, so you are denying the claim of them being true, just to keep your skin safe. Violation of my second rule (first rule accepted as existence of present human form is common, if pointed out, else even first rule is also violated) because what is common is not pointed out.

That's all human has ever done in past, to keep his (or her) own skin safe, even if it is on the cost of truth. Don't misuse analogies. They are not to confirm. They are just to compare. So, never compare a laptop with an apple. It doesn't make sense. Use analogies wisely.

Stop accepting, start thinking.

Monday, October 10, 2011

Got to have a patent now

Now this is about two weeks late that I am publishing my idea here, which I believe, must patent before anyone else does. Here is the Idea:

Creationists say that evolution is just a theory. I have got an idea which can crack their this point by a very simple, though, but a strong means. It is somewhat like LR - DI numerical (Logical Reasoning and Data Interpretation) in which you have few statements (which you assume to be true) and find a conclusion on the basis of them.

My intuition:
Assume the following statement to be true (for the time being, and I will show you how they are true in the latter part of this post)

1) A change in genetic material may cause a change in phenotype.
2) A change in genetic material may be due to natural forces.

Conclusion: A change in phenotype may be due to natural forces.

Now, if you assume my two statements to be taken as granted to be true, you arrive on the conclusion, which states that phenotype may change naturally. This change in phenotype, in other words, is referred to as variation and accumulation of such large number of variations is known as Evolution which is perfectly true.

Now the only thing left is "How do I know that the statements given are true?" For the answers, let us take these two statements one at a time.

1) A change in genetic material may cause a change in phenotype:

Please pay attention to the term 'MAY'. It says that a change in genetic material may or may not change the phenotype of an organism. Why? Because there is a lot of genetic material out there which acts just as a passive spectator and does nothing productive inside the body. And the same is true for almost all the species of organisms besides human. Still, there are sections on the genetic material (DNA or RNA) which are frequently responsible for the phenotype characters making us the way we are.
Now, it is very evident that the changes in that part of a DNA or RNA which is responsible for some sort of physical characteristics of the organism, will surely affect the phenotype of it, be it an adverse effect or a positive one. PROVED

2)  A change in genetic material may be due to natural forces.

Again the term MAY pops out. So, there has to be some sort of genetic changes that are not natural (artificial or due to human intervention) to support my statement. Of course, there are certain mechanisms by which we humans have altered the DNA sequence or RNA sequence of organisms. That is what 'Genetic Engineering' is all about. But may be all types of genetic changes are artificial. Yeah! may be, provided there are no examples where genetic material changed even if we did not want it to happen. But alas! We have ample such evidences where it happened. So, obviously, there can be a natural change in genetic material without human intervention, for example, mutation. PROVED. (Please study mutation in depth if you are not yet convinced. There might be some faulty concepts you must have built.)

Now, once we are at a stage where we can proudly say that our statements are true without any faulty assumption. Once it is so, on the basis of mathematical reasoning, we may come to the very simple conclusion of:

Conclusion: A change in phenotype may be due to natural forces.

To be noted, that I say, it MAY happen, not that it is always going to happen, all the time. So, there exists a probability, no matter how low be its value, it is never zero, and never 1. But chances are always there that an organism may change to another form,without a prior notice. It doesn't matter how small that change be (or how large, as well) which is known as variation. Variations may be somatic (that do not pass on to next generation), or germinal (that pass on to next generation). As we are talking about genetic changes, the variations we will encounter here, are highly likely to pass on to their offspring (again, it is highly likely, not always. Why? Study meiosis and crossing over in depth). But since, we made through, to prove that variations also occur naturally, we are almost there to prove that evolution is a very simple and natural process, a result of the two statements I gave in the beginning, whatever be the factors.

You see, the term MAY is always with us. And this MAY is the reason why it took so long to evolve species in geological time period. Otherwise, the amount of evolution of millions of years could have happened in just a matter of days or even hours provided there is an infinite source of resources, and infinite sink for the dead and waste (and a very fast rate of reproduction as well). But as we know, that this ideal condition could not be prepared on the limited space of earth (also consider plants to consume space and resources, not only animals), there has to be a competition for the one with best 'features to adapt' to survive, and the weaker ones are eliminated. This was what Darwin said as the 'survival of the fittest'. Hence, the theory of evolution is a highly logical, and even mathematically provable science. PROVED.

I have filled up the form for the patent of my this idea of 'statement-conclusion' method to justify evolution. By the time, I have published it here on my blog. Though I did not know that even ideas might be given a patent, yesterday I came across Kalam Cosmological Argument, and it struck me 'O my! (I don't believe in GOD, so, have to edit the phrases in accordance), I can also get my hypothesis patent, if this foolish hypothesis maker can.

Thanks for reading so far and tolerating me. Enjoy the life without the fear of god.