Dear Alexander, I would advice the Lexicology Handbook from Cruse et al (2002).

• Cruse, Alan et al. (eds.), Lexicology: An International Handbook on the Nature and Structure of Words and Vocabularies / Lexikologie: Ein internationales Handbuch zur Natur und Struktur von Wörtern und Wortschätzen, (Handbücher zur Sprach- und Kommunikationswissenschaft 21), Berlin/New York: Walter de Gruyter. 2002

Also the Wikipedia pages on Onomasiology and on Etymological Dictionaries perhaps will provide you further literature. I hope this may be helpful.

http://en.wikipedia.org/wiki/Onomasiology

http://en.wikipedia.org/wiki/Etymological_dictionary

See:

http://psychologydictionary.org/paralogia/

A clinically and forensically important dream state is the so-called Ganser syndrome, the chief symptom of which is paralogia: the patient gives a wrong answer or an answer which is so far-fetched that one can detect that the general idea contents related to the question have been stirred. This paralogia is noticeable even when

asking the simplest questions; as, for instance, "What is 2 plus 2?" "How many fingers on the right hand?" "How many eyes have you?" etc. The patient can not

name well-known objects. [Taken from: "PSYCHIATRIC - NEUROLOGIC EXAMINATION METHODS with special reference to the significance of signs and symptoms"  by Dr. August Wimmer, C. V. Mosby Company, St. Louis, 1919]

For more, see:

http://scans.library.utoronto.ca/pdf/3/5/psychiatricneuro00wimmuoft/psychiatricneuro00wimmuoft.pdf

I think Alexander should explain in what context / area of science he use the term paralogia: genetics, linguistics, philosophy or psychology? Loking at his topics and skills I think it would be linguistik or semantic.

Evasion:

Act of not facing up to or of strategically eluding something. It consists of suppressing an idea that is next in a thought series and replacing it with another idea closely related to it. Evasion is also known as paralogia and perverted logic.

http://www.newagepublishers.com/samplechapter/001696.pdf

Dear colleague, 

I am sorry for not answering in time. Technical problem.

Dear Alexander, Thanks for voting up my comment, even if this wouldn’t be necessary at a counter question. I would rather like to get an answer on my question. Although you express your regret from answering late, you do not respond to the question.

Dear Michael Krichbaum,

Thank you for your great support, You advise very interesting resources. You gave the reference of the text, this is excellent book, now I am in deep analysis of it. You were right when you guessed the idea about this question, I analyze the semantics of words that are connected with paralogia.

Dear Nebi Caka,

It was wonderful to see the resources You gave. Especially, about Ganser syndrom, link on http://scans.library.utoronto.ca/pdf/3/5/psychiatricneuro00wimmuoft/psychiatricneuro00wimmuoft.pdf was strictly to my topic. It was amazing support. Thank you. It was a real surprise that people who do not know me, help me. Really, great!!!

Dear colleagues, thank you!

Also I want to add: I use the term paralogia in sense of all spheres of science: genetics, linguistics, philosophy, psychology. Linguistics and semantics at the same time.

With Great Respect to You.

Dear Alexander,

The best dictionary about paralogia is to observe people around you . You will notice there is a continuum between what is considered normal amount of paralogia and what we clinically call pathological.

Observing people also makes you free from any frame and able to discover new forms of paralogia.

Dear Brahim,

Yes, you are right. In a certain sense, it is the best way to find out some deviations. But it is not possible to identify person as deviant from view....

Alexander,

I understand what you mean...I just think it is feasible if you choose an n people randomly than observe them without their knowledge. Make your own scale from 0 to 10 for example and see what you can find. Then you can compare your results with another researcher who does the same thing over the same n people. I will be very interested to see the results. I mean do not you think it is important to statistically measure how many pathological cases we are talking about. How many people are on the verge of deviance , how many are really deviant etc...

But you are right-I totally agree with you ..