The scientific method involves testing of a hypothesis by an experiment. If the experiment shows a result that supports the hypothesis, this is considered as positive result, worthy of publication. If the experiment does not show the expected result, this indicates that the hypothesis was incorrect. This sort of “failed” experiment is called a negative result and most journals won’t accept such results for publication. Should it be published or not?
In my understanding, you need to publish all the results in the way of finding the right solution. This is even very useful to see the entire chronology of the process, because in science a negative result is also a result, because tracing this chronology gives a better understanding of the process of finding a solution by other people. This is very important for scientists, but not for editors of journals, because it gives the opportunity to see the creative laboratory of finding solutions from all sides.
I want to quote one of answers from the discussion
https://www.researchgate.net/post/What_is_your_vision_of_the_failure_of_the_researcher_for_the_first_time_in_his_research_experience_Do_you_mean_stop_or_what
"The researcher should expect to fail several times before finding success. As an inventor, Thomas Edison made 10,000 unsuccessful attempts at inventing the light bulb. When a reporter asked, "How did it feel to fail 10,000 times?" he famously answered,
"I have not failed; I've just found 10,000 ways that don't work".
In my understanding, you need to publish all the results in the way of finding the right solution. This is even very useful to see the entire chronology of the process, because in science a negative result is also a result, because tracing this chronology gives a better understanding of the process of finding a solution by other people. This is very important for scientists, but not for editors of journals, because it gives the opportunity to see the creative laboratory of finding solutions from all sides.
I want to quote one of answers from the discussion
https://www.researchgate.net/post/What_is_your_vision_of_the_failure_of_the_researcher_for_the_first_time_in_his_research_experience_Do_you_mean_stop_or_what
"The researcher should expect to fail several times before finding success. As an inventor, Thomas Edison made 10,000 unsuccessful attempts at inventing the light bulb. When a reporter asked, "How did it feel to fail 10,000 times?" he famously answered,
"I have not failed; I've just found 10,000 ways that don't work".
Not if does not bring us further to a solution. Indeed, a failed attempt can be very enlightening, if there is new theory developed along the way - that theory, if it stands on its own, could be part of a future solution to the problem - or indeed another, related, problem. But - I iterate - a paper must have a substantial value on its own - failed attempts that does not lead us further on shall not be published.
I certainly agree with Gennady Fedulov . In fact, an experimental outcome that does not support a hypothesis does not mean the hypothesis itself is incorrect. It just means the results did not provide enough evidence to enable the researcher to accept the hypothesis. This in itself is not a bad thing. It just shows what doesn't work! And in my opinion, this outcome also advances or consolidates our knowledge on what works and what doesn't work.
As to whether a journal will accept this result depends, to some extent, on how the researcher makes their case in the light of previous and future work. I think negative results should be given a fair shot at publishing!!
The famous Russian mathematician Grigori Perelman published his three papers to prove the Poincaré Hypothesis not in a prestigious high-ranking journal, but on the Internet. Nevertheless, it was enough to recognize his work as a great event. Perelman have been spitting on journals and reviewers, he proved this result to satisfy his personal ambitions without any intentions to be rewarded. Yes, this is a unique case. Then in the USA two independent groups of expert mathematicians were created who tried to find a mistake in the proof of Perelman, but could not. In the end, the proof was recognized true and Perelman became famous. Of course, Perelman was not 100% sure that there was no error in the proof. His position was flawless: if the error were found by someone, it would bring glory to the one who found this error. It is necessary to note one very important thing: this is the very case when the method of proof is much more important than the result itself, despite the fact that this result was worth one million dollars, from which Perelman kindly refused.
I am of the opinion that negative results should be published as long as it is a pointer to the other side of the coin. The most important thing for both the so-called positive and negative results is "Did the process of investigation follow the right scientific method in the shades of ethics?"
One other problem with results is the ability of the researcher to convince the reader on the applicability of the findings and the link-ability of the findings to previous studies and theories. You need to convince the reader or reviewer on the outcome of the study whether positive or negative.
For the case of being accepted by editors for publication or not, I think if you have a strong point, the points to earn for publishing in a journal should not be your concern (as the scientist). You can use other means as Gennady Fedulov cited in his example of The famous Russian mathematician Grigori Perelman.
A scientist should be open-minded and objective.
Negative scientific results have to become publicly available from two main reasons: a) research costs, b) effectively of science.
a) Research costs are dramatically increasing not only within states but globally as many research projects are unnecessary being researched over and over. A publicly available database of researched themes and the negative results can help to save a lot of money.
b) Effectively of science is hampered by a multiplicity of the identical research projects. Database of all research attempts will enhance out the speed of gaining new knowledge.
c) Database of all researched projects with their outputs can help to build a new metalevel-science working above those already researched projects. It can become a base for a strategic scientific planning of novel, promising research directions!
d) Personally, I do see future in automated, using AI, research at the metalevel of the science. I just see a possibility to build a strategic software that will help to navigate humanity towards promising scientific areas and strategies that are not obvious due to lack of us, humans, to keep huge amounts of information simultaneously in our heads.
It is obvious that if we have an outstanding conjecture on an important problem , then it is important to settle whether the conjecture is true or false. But if the conjecture is of minor importance (and how to determine if it is might be tricky, and it also changes over time!), then we should let it be.
- Badges
- Science topic