I consider original scientific research to be the primary source.  I think journals take the same view.

'Original research is considered a primary source.  An article is considered original research if...

• it is the report of a study written by the researchers who actually did the study.
• the researchers describe their hypothesis or research question and the purpose of the study.
• the researchers detail their research methods.
• the results of the research are reported.
• the researchers interpret their results and discuss possible implications.

http://libguides.unf.edu/originalresearch

"What in your eyes is original scientific research?"

I really do not know, and I think it is to difficult to answer nowadays in our technocratic world...

eventually even the definition of science is different to its classical definition, and the limits between science and engineering are more diffuse than ever

"How far, in your eyes, is such a claim true?"

Compared to my understanding of classical science and in my field of visual sciences overall "close to zero" with very notable and outstanding exceptions!

The problem is to dig deep enough in the literature to find those notable contributions among the ocean of incremental science

In my other field optics, the situation is a bit different but because in optics there is aso very much engineering (with great merits) deserving publication, but there the proportion of truly original contributions is much higher (although in both fields revisiting topics and reviews are very often)

"How exactly does a journal define originality of researches?"

I understand in general:

original as "never published before" and not constituting "incremental science"

but this is also very vague

I consider original scientific research to be the primary source.  I think journals take the same view.

'Original research is considered a primary source.  An article is considered original research if...

• it is the report of a study written by the researchers who actually did the study.
• the researchers describe their hypothesis or research question and the purpose of the study.
• the researchers detail their research methods.
• the results of the research are reported.
• the researchers interpret their results and discuss possible implications.

http://libguides.unf.edu/originalresearch

According to WIKI, "original research is research that is not exclusively based on a summary, review or synthesis of earlier publications on the subject of research. This material is of a primary source character. The purpose of the original research is to produce new knowledge, rather than to present the existing knowledge in a new form (e.g., summarized or classified).... Original scientific research published for the first time in scientific journals is called the primary literature. Patents and technical reports, for minor research results and engineering and design work (including computer software), can also be considered primary literature. "More reading follow!

The third paper is good for @Miranda! I think original scientific research which was published is treated as primary resource! It is absolutely new research that brings absolutely new knowledge!

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

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

http://wac.colostate.edu/books/thinkingwriting/chapter6.pdf

Is primary resource the only point to determine whether researches are original? What about mathematics? In mathematics, there is no question of primary resources. 

Dear @Hemanta, originality is the key issue but please read the following: "... Originality can be described as something that has not been conceived of or done before by someone. With a vast body of research literature out there, how are we to know what we’re doing is original research? How do we know we’re not duplicating someone else’s published work? Short answer: it’s almost impossible to know for certain. There’s an apocryphal anecdote that John von Neumann came up with two solutions to two problems, but on both occasions afterward learnt that someone had already solved the problems. Another example is Gauss’ work. Gauss didn’t publish much of his mathematics research. In many cases, later works were found to have been rediscoveries of Gauss’ unpublished work. Or consider the calculus war, a priority dispute over whether Leibniz or Newton was the original inventor of calculus. One of our best guards against duplicating someone else’s work is to be aware of the general research trends in our chosen area and have a clear statement of our research questions. These two topics will be discussed in turn...."

http://mvngu.wordpress.com/2011/07/22/originality-in-research/

Dear Professor Jacic,

Oh! The times! Oh! The manners!

The days of Newton and Gauss are over. Currently, scientific temperament is hardly seen among researchers. How, do you think, originality would be recognized these days?

In fact, Professor Jacic, we think in company only when there happens to be something to think about uncovered by someone who had thought about it individually. The person who thought about something for the first time is the original thinker; the others who follow the footsteps of the original thinker think in company. 

Dear @Hemanta, originality refers to the ability to think independently and creatively! According to WIKI, "Originality is the aspect of created or invented works by as being new or novel, and thus can be distinguished from reproductions, clones, forgeries, or derivative works. An original work is one not received from others nor one copied from or based upon the work of others. It is a work created with a unique style and substance."

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

Yes, Professor Jacic, and to think independently, one must think alone, and not in company.