A hypothesis is a tentative proposition or posit based on insufficient knowlege to be sure that it is factual. A hypothesis is proposed for testing.

If much testing affirms the correctness of a hypothesis, and it is generaly accepted, it then can become accepted as a theory. However, theorys can still be challenged and they may be modified or even discarded altogether, if much contrary knowledge is acquired and presented.

If a theory is rock-solid and apparently is beyond any dispute, it can be accepted as a law. There are laws in physics, for example. However laws are very scarse, or non-exsitent, in other diciplines such as biology.  

Paradigms  are also interesting, if you are keen. They are, very roughly, generaly accepted principles within which research is coducted, but they may be overthrown and replaced by a new paradigm during a scientific revolution.

Note that in non-scientific language, in common speech, even an idea or a train of thought may commonly be referred to as a 'theory', and the word hypothesis is not generaly known or used, and law is usualy used only to refer to the legal system.

I hope this helps Alex,

Regards,

Keith

The number of unjustified hypotheses (not supported by any theory) or un-testable hypotheses, is infinite. Therefore it is useless to propose and test such hypotheses. Yet, in the routine experimentations, such hypotheses are sometimes tested with a hope that some discovery will happen by accident.

Almost all basic books that treat philosophy of science  treat the question of hypotheses and theories. Some more basic or older articles may be more clear. The book, that must be read, is of course Popper’s “The logic of scientific discovery”.

http://strangebeautiful.com/other-texts/popper-logic-scientific-discovery.pdf

 A hypothesis is based on observation. The observation may be in conformance with a theory or counter to a theory. Testing an observation. The observation may be that X happens when Y happens. There may be no causation between X and Y. The observation is that X happens when Y happens. It is not necessary that X always happens when Y happens or that X cannot happen without Y. Nevertheless, it is necessary to state the expected, testable relationship of X and Y or explain that a relationship is sought. There is no  need to perform null hypothesis testing.

Hypotheses are on one hand always based on theories, because language itself contains theorie as well as perception and observation does. The questions are: Is a theory explicit or implicit? Is a theory consequently used? Is it consistent? Is it deep or superficial? Is it clearly formulated or vague?

One the other hand there are a lot of problems with the deductive idea that every empirically tested hypothesis should derive from a theory (e.g. Karl Popper), because sicence is empircally often enriched by sudden observations or findings by chance (e.g. Ian Hacking).

In order to plan research projects I wold say it is better to build it on an explicit, clear, consistent and successive theory. But it is hard to decide if induction or deduction is a better guide to scientific progress. To me it seems that the truth is somewhere in between.

Regards Thomas

Hypotheses are on one hand always based on theories, because language itself contains theorie as well as perception and observation does. The questions are: Is a theory explicit or implicit? Is a theory consequently used? Is it consistent? Is it deep or superficial? Is it clearly formulated or vague?

One the other hand there are a lot of problems with the deductive idea that every empirically tested hypothesis should derive from a theory (e.g. Karl Popper), because sicence is empircally often enriched by sudden observations or findings by chance (e.g. Ian Hacking).

In order to plan research projects I wold say it is better to build it on an explicit, clear, consistent and successive theory. But it is hard to decide if induction or deduction is a better guide to scientific progress. To me it seems that the truth is somewhere in between.

Regards Thomas

I do not really understand what Joseph wanted to say. Let me continue anyway.

A hypothesis (which normally has the same logical structure as a theory), that is based on a theory - recognized as reliable theory, may be justified, but this depend how strongly the theory is justified itself. And, alternatively,  of course, if a hypothesis is not based on any theory whatsoever, such hypothesis is just a guess. Like tossing a coin: you chose a hypothesis just by chance. The probability that such a hypothesis will be later proved to be true is almost zero. In research, we never propose hypotheses by tossing a coin. It is true, there are some extreme cases when some hypotheses are proposed without much theory behind, but are chosen intuitively, because the researchers get some feeling or have unconscious knowledge of the field where they work and the hypotheses that are “not based on theory” in fact ARE based on some unarticulated, unconscious theory. Some very experienced researchers may afford such reasoning… sometimes (Edison, Tesla, Einstein). May be this is what you have in mind? In conclusion, I must say this: hypothesis that is unjustified has practically no value at all.

Indeed, there are people who propose that everything in the world is connected, so that every single event or entity is connected to all other entities or events or facts. So, if you follow such logic, there is no hypothesis that is not based on some theory (known or unknown theory). But I think that this is crazy.

Theory is generally considered an established line of thought and explanation. A hypothesis is a logical, testable statement concerning an observation. The meaning of theory and hypothesis are often muddled by language usage. To have a theory about something may be the same as having a hypothesis concerning an observation. Both are a preliminary explanation. A scientist must understand the causative factors in the test of a hypothesis in order to properly test the hypothesis. Science is not a matter of tossing dice. An observation is made and an explanation is offered. The explanation is used to perform the test based on a logical statement, if Y then X, no X without Y, or usually X after Y. You might call the explanation a theory and the logical statement a hypothesis. Indeed, the explanation may be an established theory. Nevertheless, calling the postulated explanation a theory is incorrect. An explanation and a logical statement is the hypothesis.

The scientific method was designed to exempt scientific inquiry from authority. Authority may be an institution, a person, or an established theory. Popper's ideas on science often clash with the scientific method, but they are not wrong by way of clashing. Popper's most important statement is that the hypothesis must be falsifiable. A falsifiable hypothesis is a complete logical statement with no ambiguities. Ambiguity leads to a non-falsifiable hypothesis.

From the question

"...(T)he authors argue that X (an action) should be related to Y (an emotion). In order to support this argument the authors suggest that when individuals engage in X, they are more likely to feel a sense of absorption and thus they should experience Y. There is no theory here to support the relationship between X and Y."

Change the word 'theory' in the last sentence to 'explanation,' 'method,' or 'mechanism' and meaning is the same. The authors rely on a 'sense of absorption' to lead to an emotion. The supposed explanation is the 'sense of absorption.' Failure to define sense of absorption and lack of a proposal to measure sense of absorption results in a non-falsifiable hypothesis. Does X restrict the types of actions and does Y restrict the type of emotion? If not, the hypothesis is non-falsifiable.

A common error occurs when constructing hypotheses that rely on indirect evidence. The error is the construction of the hypothesis around the indirect measurement instead of the effect indicated by the measurements. Many physical experiments cannot measure the effect directly. There must be a defined mechanism from A to B to the measurement C. Here, the quantity or effect B is the desired result that must be interpreted from C. An explanation of A to B to C is an essential part of the hypothesis. The softer sciences also rely on indirect measurements in many situations. If, say, the experiment in the question measured Y, an emotion, to estimate a sense of absorption, then a proposed mechanism, method, or explanation is essential. The postulated mechanism, method, or explanation is not a theory, but is no less essential.

A hypothesis may not be open-ended (non-falsifiable). The path to the conclusion must be clearly defined. The conclusion may not be a surprise. When a surprise occurs in the course of an experiment, the surprise is a new observation. If the observation is important, then a new experiment is required with its own, falsifiable hypothesis.

The problem with theories and consequently with hypotheses has to do with the act of knowing. As a matter of fact a theory is nothing but the synthesis made of the thing perceived through the sense and the thing added from the intellect by means of imagination (Kant). The same as the act of knowing, a theory must be based on the synthetic connection of intuition. A theory must define its object of study and determine  the facts constituting the theory (Coseriu).

Science and theories are constituted with statements. Science starts with an intuition, made into a statement and this one into an assumption thus adding necessity and universality to it and ends in a series of statements deduced from the truth in the initial intuition cosntituted in the initial assumption. A theory must be adequate to the object it studies. It must be certain in terms of the degree of reality of the statements constituting the theory (Ortega y Gasset) and the internal logic of the theory (Popper). And it must be true if facts can be verified in the real.  With this the theory consists in something verified in experience because a particular fact was constituted as an image of itsef, that is, it was made universal and necessary (Descartes, something added in the formulation of the theory), thus representing the class the objects to be studied belong to, and because the concepts representing the object of study and the facts determined were translated into statements of a language made a specific language (a terminology or nomenclature, Coseriu). From this point of view a hypothesis is nothing but a fact not yet verified (Martínez del Castillo). The same as with other facts depending on a theory, which are verifiable and can be verified once and again, a hypothesis, due to the individual character of the fact constituting the hypothesis, can not be verified, but it will sooner or later, because the theory so far has proved it to be adequate, certain and real and the fact to be verified is based on analogy (Ortega y Gasset) with ones already verified in the real. So a hypothesis is nothing unless there is a theory stating the universality and necessity of the thing being studied, the object of study of the theory, and the character of the facts having being deterrmined as facts within the theory. Becuase the facts determined by the theory as facts are true because they have already verified in the real, the one constituting the hypothesis must be real and true.

But as we have seen, in theories language is present. And together with language there is a community of speakers who think together in a particular mode of thinking constituted in a community of thought (Martínez del Castillo). Scientists necessarily belong to a particular community of thought. All speech communities transmit particular modes of thinking based on beliefs (Ortega y Gasset). Ideas transmited and accepted by speakers from their childhood are considered as if they were the real things. Because of this it is necessary to analyse the statements a scientist starts with thus looking for the radical reality they are based on (Ortega y Gasset) or the internal logic of the theory (Popper) and once the problem of demarcation has been solved (Popper).

https://www.researchgate.net/profile/Jesus_Martinez_Del_Castillo2/publications

http://strangebeautiful.com/other-texts/popper-logic-scientific-discovery.pdf

Book Modes of Thinking, Language and Linguistics

Article International Journal of Language and Linguistics Modes of T...

Book Psicología, lenguaje y libertad

Joseph and Jesús,

My God, please help me.

Those whose kingdom is in the Haven would like just to know "Should hypotheses always be based on a theory?" Please help them.

Thanks very much indeed.

Hypotheses will not be hypotheses unless they are based on a previous knowledge. On the contrary hypotheses will constitute an un-verified interpretion of a supposed-to-be truth. In hypotheses the scientist must look for the confirmation of them. if the scientist can verify a hypothesis, the hypothesis means something. It is consititued with some previous knowledge. This previous knowledge is what we usaully call an assumption. It is necessary to find out the terms the hypothesis is based on. If the hypothesis in question is veritied, that hypothesis was not a hypothesis but an assumption.  If a hypothesis cannot be verified it is not properly formulated. This problem has to do with the problem of demarcation (Popper) and a theory of knowledge. I studied this problem in several works of mine. I gave examples of certain hypotheses having been formultated in the study of language and the reason and consequences of them. The problem is in analysing the statements the hypothesis is based on. All hypotheses are based on previous knowledge thus constituting the assumption they are based on, that is, the theory. See the article and the book below.

Kind regards

Jesús

Book Psicología, lenguaje y libertad

Hypothesis A suggested explanation for an observable phenomenon or prediction of a possible causal correlation among multiple phenomena.

Theory A well-substantiated, unifying explanation for a set of verified, proven hypotheses.

No. A hypothesis does not require a theory. An explanation and a logical statement is the hypothesis.

Joseph, jésus,

As I thought.

Sorry. I do not agree. The reason is that we need a justification for a proposed hypothesis. A justification is already a theory. In fact Jesus admits this also. See my first comment on this thread. 

Dragan

Justification is not a theory. You are confusing terminology. It was necessary to clear up the situation after reading your first comment. Hypotheses and definitions are not derived from "Alice in Wonderland," but from logical statements.

Joseph,

Please read your Gamut with more attention.

Refer, also, for examples about theory, to what Popper cites as "theory". An example that I give often is: any simple human action or even movement is based on theory about the world.

On the other hand the logical forms of a hypothesis and theory are the same.

A hypothesis may be unfounded, but as such it would be a pure guess, as I said above. In addition, number of possible such guesses is unlimited and therefore such strategy would be absurd in research.

Of course, Dragan, a justaification is, that is, acts as a theory. Hypotheses must be based on some knowledge previous to the formulation of the hyphothesis. You know something about the topic the hypothesis deals with and must verify the rest you do not know.  That is the sense of the hypothesis.

A hypothesis is a tentative proposition or posit based on insufficient knowlege to be sure that it is factual. A hypothesis is proposed for testing.

If much testing affirms the correctness of a hypothesis, and it is generaly accepted, it then can become accepted as a theory. However, theorys can still be challenged and they may be modified or even discarded altogether, if much contrary knowledge is acquired and presented.

If a theory is rock-solid and apparently is beyond any dispute, it can be accepted as a law. There are laws in physics, for example. However laws are very scarse, or non-exsitent, in other diciplines such as biology.  

Paradigms  are also interesting, if you are keen. They are, very roughly, generaly accepted principles within which research is coducted, but they may be overthrown and replaced by a new paradigm during a scientific revolution.

Note that in non-scientific language, in common speech, even an idea or a train of thought may commonly be referred to as a 'theory', and the word hypothesis is not generaly known or used, and law is usualy used only to refer to the legal system.

I hope this helps Alex,

Regards,

Keith

I have always considered hypotheses and theories related just as Keith so clearly stated. Viewed in this way you cannot have a 'theory driven hypothesis'. If you can find a theory which fulfils this requirement and is applicable to human executive functioning as you describe you are doing very well indeed. I suspect there are few hypotheses in this area worth the name, let alone theories. Suggest you read, in order, Popper, Kuhn, Lakatos and finally Feyerabend for their views on theories.

Restriction of the scope of hypotheses so they accord with existing theories is in my view artificially constraining what one may or may not ask about nature. I feel that few major scientific discoveries have been made on the basis of theory alone (maybe confirmation of the Higgs boson would count as such). I think that is a case of the exception proving the rule. So in your example don't worry too much about theory - if you can produce a self consistent description of the observed processes, regardless of whether they accord with existing thought or not, then I think you have made progress, and I wish you well in your endeavours!

Tony

All "confirmational" experiments have a theory driven hypotheses (see verification, meaning verification of some theory). Normal routine experiments are of this kind.

The so called "crucial experiments" also. Those later are experiments that test a critical section of some theory, so they are also based on some theory. In linguistics for example if we would have a method that can demonstrate the brain structures that correspond to "generative grammar" would be a crucial experiment that would, if the structures were found, confirm the hypothesis of generative grammar.

A third kind of experiments, for which we propose a new hypothesis based on some "other" maybe unrelated theory, may look like that they do not have well founded hypothesis, but in fact they have one, although sometimes shaky hypothesis. Mickelson-Morley experiment probably was one such experiment.

Only when a hypothesis is pure guess, without any either correlation or association to the nature of the event tested, this could be a hypothesis where there is no theory behind. As I said such hypotheses are only accidentally useful, otherwise they are absurd.

Thank you everyone for your insight.

I think the issue is being complicated more than necessary. Simply stated, a "theory driven hypothesis" is one provided when you have some idea regarding the relationship among certain variables - it is a priori, and the purpose is to either confirm or refute what you predict will occur; data driven hypotheses, on the other hand, are those provided when the goal is to provide a plausible explanation as to how certain variables are associated, and are done post hoc. In a sense, data driven hypotheses are used to derive theory, while theory driven hypotheses are used to test theory.

I think Keith and Tony have very well explained the relationship/connection between hypothesis and theory. And the issues and constraints on them. I think it is while inquiring into the human mind and consciousness as phenomena in Nature that one confronts the problems and limitations of the existing processes of hypothesizing and theorizing, which have largely been used and developed in relation to external phenomena. But this also challenges us to rethink and bring new inputs into our existing methods of hypothesizing and theorizing. In the end i would just say that the purpose of using these processes should always be remain primary within our minds and all hypothesizing and theorizing must always remain subordinate to that. And that aim broadly speaking is to keep reducing the gap between the objective reality and logic of any phenomenon/process that we are trying to understand and our idea reality (which we construct through hypothesizing and theorizing) about it.            

Sadia, Tony,

Well, guys, try to get a grant with a project where the research hypothesis is not "well grounded" (has a lot of theoretical support). Or, for example, write a paper that shows that the Japanese poetry has or does not have an effect on the frequency of an iron pendulum hanging in your lab - and try to publish such a paper.

Joseph L Alvarez's , Jesus Martinez and other submissions by researchers have answered the question in detail. A hypothesis is a suggested solution for an unexplained occurrence that does not or does fit into current accepted scientific theory, or that seeks to substantiate an observation which fits into a theory. I support the fact that,  the observation will certainly have a theoretical underpinning. Hypothesis should connote to scientific theory for it to be testable or falsifiable. The basic idea of a hypothesis is that there is no pre-determined outcome. For a hypothesis to be termed a scientific hypothesis, it has to be something that can be supported or refuted through carefully crafted experimentation or observation.  

This is a good description of the position of orthodox science on hypotheses, theories and experiments, which I have little problem with, excepting the relative priorities of theories and hypotheses. I see them more as observation -> hypothesis -> theory, as Keith stated.

What I do take issue with however, is the implied assertion by science that this is the only way to make progress and widen the field of human knowledge. Application of the scientific method, by design and definition, adds to the human knowledge base in small steps. In many cases large steps have been made in other ways, such as Kekule's visualisation of benzene type molecules as rings, Einstein's mass energy relation equation, Susskind's conception of the universe as a hologram and so on. None of these were based on any sort of theory, or so I understand.

I totally agree with the comments about attracting research grants. But let us get the causality correct. Do research grants get awarded for theory based projects or do they define what is allowable as a theory based project? Personally I believe the latter is closer to what actually happens.

I would encourage all up-and-coming researchers to always bear in mind that they may be on the verge of a major breakthrough, and not to let themselves be straight jacketed by existing theories, of which the only thing one can incontrovertibly say is that one day they will be proven, at least partially, wrong. The sad part is, that researchers who overdo this will not get the academic recognition and higher degrees that they may deserve. The plus is that we will almost certainly need every bit of thinking 'outside the box' of which humanity is capable both to understand quantum theory and unify it with general relativity.

David

It may be based on intuition, feelings, preferences like colors of the substance, package, or eyes of the person who asks the question, on how her name sounds, free associations, random numbers, date, dreams, desires. Anything. You of course will not get your PhD if you would proceed like this, except if some miracle would intervene, but then you may get a Nobel prize for guesswork. But if PhD matters to you, chose one hypotheses that makes sense with at least some minimal theory that you can put behind it.

David, Tony

We must take hypotheses (as theories) as propositions that are very pure in content. When I move my hand, my theory about the world predicts its position in the next instant, and so on.The "minimal theories" in science (that I mentioned above, that would justify hypotheses) may be not much more elaborated and may be based on intuition (that is not completely blind for the theories!), correlation thinking, associations (see Paul Tagart) or even resemblance thinking. Or of course there may be full blown theories, as I mentioned above, that are behind scientific hypotheses.

Tony, no theory predicts all possible outcomes and is therefore not so devoid of creativity. "Verification" consist in fact in verifying such theories. However, if the outcome is certain, there is no need to verify a theory. bad research often tests the outcomes that are certain or serves some security purposes. Testing such theories also provides better and more “bold” formulation of a theory (as Popper would say) and makes them more susceptible to refutation.

Notwithstanding some of the good comments (and some of the not-so-good comments) that have already been made, Alex's question seems to me to be a badly framed one since there is—as one can see from the comments of others—no clear, genarally accepted account of either what a hypothesis is or what a theory is. As these terms are bandied about (not necessarily without point), many things that one might naturally call a hypothesis might also be referred to as a theory—even by the same person in the same sentence. In a context where someone has made clear what s/he means by "hypothesis" and by "theory", a question like Alex's might profitably be raised. I doubt that this is the situation in the paper about which she frets. A number of the commentators here make declarations about what is meant by "hypothesis" and/or "theory". Where accounts are offered about what might be meant by these terms in a particular context of discussion, in order that someone might get somewhere, they may be all to the good. However, if they pretend to be explanations of what these terms "really" mean "in science", or in the philosophy of science, they are not helpful, as no such common understanding has been achieved. Both "in science" and in the philosophy of science, as well as in commonsense discourse, "hypothesis" and "theory" are polysemous terms. What one means in using them may be clear from the context—or not. Alex would do well to ask herself what she is really asking about; evidently, the authors she cites don't talk explicitly about either "hypothses" or "theories". According to Alex's brief account, they merely propose—and here one might say here "hypothesize" or "theorize"— that an action is related (necessarily?, typically?, characteristically?, usually?, in certain kinds of cases?—related to (?) an emotion. Alex says that they argue for this (whatever the "this" is here, which is hard to divine). Well, if so, what Alex should be concerned with is (a) what they actually mean, which from her description is pretty difficult to divine, and (b) whether their arguments are any good: not whether their hypothesis/theory should be based upon a theory/hypothesis (whatever Alex may mean in using those terms, which is also hard to divine).

A hypothesis seems to usualy emerge in someones mind due to either empirical finds, a synthesis of theoretical considereations, or a synthesis of  empirical findings with theoretical considerations, and can be derived by either deduction or induction. New hypotheses should abound in biology now, due to the massive surge in the   available and very surprising empirical data  that have come from the recent sequencing of many genomes. 

Hypotheses are derived for testing, and can be either discarded, modified, or accepted,  after such testing. This brief note should be read together with my previous answer, to get a fuller  picture of my comments on the posed question.

Nice discussion so far.

Let me add some additional arguments. The necessity to build hypotheses on theories to me seems only relevant if hypotheses are additionally understood as interlink between an empirical relation and a linguistic relation. Hypotheses are not only justified by their relation to theorie. On one hand they are even more than theorie, despite they often use less words. Most theories contain unprovable aspects and preassumptions. If one argues that hypotheses only have to be deduced from theory it implies that all the preassumptions and unprovable aspects influence the hypotheses. And an other aspect seems additionally relevant: A concrete hypotheses could probably be deduced from different theories.

Scientific hypotheses must fulfill much more criteria than their association with any theory. Their linguistic form and meaning should have a clear relation to empirical events. In this context there is a long debate in the philosophy of science since Empirism and Positivism, especially in analytical philosophy. The question is: How is it possible to represent experience or 'reality' by formal systems like mathematics and language. It often seems like a kind of miracle that letters, words, sentences, numbers mathematical formulae written on paper represent such complex things like the universe, the body or the mind and their functions.

Most theories in natural science are a mix of both linguistic and mathematical formulations, but in rather soft sciences the linguistic formulations are often dominant. Letting aside the debate if mathematics is a kind of very formal language or not, we have to admit that the problems with meaning of language are huge. According to Quine, if we translate assertions into an other language we often will not get the exactly same meaning, because languages are contextual systems. Therefore the problems of language based theories in science are complex. A theory could not be seen as a concrete and fully definable language independent system of concepts and relations. They could have different implications and meanings in different languages and probably lead to different hypotheses.

A good hypothesis is formulated as a relation that is testable, which means that it could be operationalized or translated into a methodological form as a kind of instruction for the researchers behavior and the (assumed) adequate procedures. And we have to admit that these methodologies themselves are based on theorie or trends. Kuhn called these trends paradigms, matrices. Some scientific theories are strong related to praradigms. In consequence they rise and fall with them. But some hypotheses are less associated (e.g. elliptic form of the plantes movements). And a complication in rather soft sciences is that a lot of different theories are used as explanation for the same observation.

In consequence there is no isomorphic connection between theory and hypothesis, between language and theory and between hypothesis and operationalization or between paradigm and research.

Nevertheless hypotheses are always in a theoretical context. There is no escape from theory. The best we can do is to be aware of the used theories and the associated problems.

Regards Thomas

David, Thomas,

No, I mean there is no relevant theory in the background. There are occasions where hypotheses do not have any "relevant" theory behind. For example if you would use random numbers to decide on a hypothesis. When a laboratory conducts tests of the effects of a line of substances on some biological process and if the choice of the substances tested is such that, for example: we test all substances which name starts with “w”. Only theory behind may be then that in the history of science we can find the accidental discoveries that were not backed by any theory at that time. Theory behind is then that “accidental experiments may produce discoveries”. I would not call such “theories” relevant theories.

Dragan,

I agree. Sometimes interesting ideas, hypotheses an results derive from sudden observation, as I stated in my first answer. And your contribution of the testing of substances is a good example for systematic observation studies. Also the finding of the microwave backgroung radiation is a good example for more or less sudden results. According to your argumentation is the research for extrasolar planets. But one could argue that behind theese systematic observational approaches there is still theory; weak expectations or methodological implications.

Regards Thomas

Yes, Thomas.

Yes, hypotheses must always be based on a theory or an idea (systematically or unsystematically) ------an X (an action) from clear or vague Y (an emotion).

Hypothesis need not necessarily based on an existing a priori theory. Results of serendipity, which later enable us to write a theory are examples. The principle of buoyancy observed by Archimedes was not obtained from an existing theory but  suddenly. The hypothesis on the shape of space to cause gravity was not based on an apriori theory but a sudden critical observation and trying to create a hypothesis that describe what was observed. Hypothesis within an existing theory is to augment an existing body of knowledge from inside while creating a new result based on a new hypothesis is introducing new knowledge from outside.     

Dejenie

Archimedes did not propose that the volume of a body can be measured by absolutely unrelated events or state of affairs, but was based on a theory he already had about water, bodies, space, weight and other properties of matter or the physical world. I mean we project such “mini“ theories all the time to be able to “predict” the next state of our mind and body. If at each instant the world that we would have to face in the immediate future, would be absolutely unpredictable, we would not be able to move.

Dejenr

This is Popper's concept in fact, but I just insist that the theories go even further, as deep as our basic understanding of our world.

Dragan,

Indeed you are right in that sense that he might have some theoretical and practical knowledge on such issues, but the discovery of that feeling in that particular instance was sudden, with out plan and with out making a hypothesis of his own from an existing theory but he was doing something else at that time, actually he was working on an assignment from local king to check out whether a gold is true or fake.  

Dejenie

Interesting example. Yet, the accidental discoveries are most often not based on hypotheses. My finding a diamond in my pocket would be one pleasant discovery. Yet if I would remember that I paid a visit to a local jewelry shop, I might start making some hypotheses about it, projecting various abductions about the circumstances under which I acquired this precious stone: obviously projecting various abductive theories.

Archimedes, on the contrary, probably did not arrive at the solution by pure accident. Plutarch maintains that he kept just making various hypotheses about everything, even when taking bath or during the very siege of the Syracuse that coasted him his life. The theory there must have been that if the two objects had the same weight but differed in volume, they certainly were not made of the same substance. The rest must have been conjectures. All of this must have been based on number of theories about the world.

Dejenie

Here is Fleming's paper:

http://202.114.65.51/fzjx/wsw/newindex/wswfzjs/pdf/21-3flemingchainabraham.pdf

It is obvious that, although facing an accidental observation, he did not proceed without a clear hypothesis that was well founded on a precise theory about bactericidal effects of some agents.

Dragan,

Yes, particularly on the second publications he pursued  his observation with a clear hypothesis and expectations.

Hypotheses may also be based on intuition, that's how the human thinking proceeds. Artists do not believe in theories, they find everything out themselves in collaboration with the public.

For me, hypothesis should be an intelligent guess about what is expected to happen or be the outcome of a research. Now what do you do in the case there are no existing theories related to your school of thought being studied. I think the best thin to do is to look for something closest to a theory that led your mind to believe your hypothesis should be an intelligent guess. Also note that both theories and hypothesis are not cast in stone. They are views that can be refuted or better confirmed at varying times.

Rita,

Please see my second comment on this thread:

“A hypothesis (which normally has the same logical structure as a theory), that is based on a theory - recognized as reliable theory, may be justified, but this depend how strongly the theory is justified itself. And, alternatively, of course, if a hypothesis is not based on any theory whatsoever, such hypothesis is just a guess. Like tossing a coin: you chose a hypothesis just by chance. The probability that such a hypothesis will be later proved to be true is almost zero. In research, we never propose hypotheses by tossing a coin. It is true, there are some extreme cases when some hypotheses are proposed without much theory behind, but are chosen intuitively, because the researchers get some feeling or have unconscious knowledge of the field where they work and the hypotheses that are “not based on theory” in fact ARE based on some unarticulated, unconscious theory. Some very experienced researchers may afford such reasoning… sometimes (Edison, Tesla, Einstein).”

The question that remains is what you refer to when you say "intuition"? What Kant had in mind or what Croce had in mind or else? Thos are completely different concepts. I presume that intuition is a concept that may be of different origin: reflex, random, associations, correlations, emotional states, intentions, remembered, unconscious tension of any kind, dreams, anything. Such hypotheses may be scientifically quite unreliable.

In art, on the contrary, since they may be the very essence of art, the intuitions are important. But - this has little to do with the question here.. For the links of art with science we may need some other thread of discussions. Otherwise we will finish, as number of other discussions finished – in despair. Anyway, I try to be an “artist” all day long, as I open my eyes.... Just from time to time, I try to be a bit of a scientist, and more concretely, on this site. So, we should keep art away from here, Rita, please.

In my personal view, not necessary hypotheses always be based on a theory.  Since hypotheses are derived from a researcher's current testable research model / conceptual framework which is underpinning on certain theory, hence hypotheses are related to theory.  However, when theory search has been exhausted after literature review or based on certain new phenomena which doesn't have any precedent of theoretical framework to support, perhaps hypotheses can be formed for testing based on the conjectured relationships.

Dear Mr. Han Ping Fung

I think at the time “when theory search has been exhausted after literature review or based on certain new phenomena which doesn't have any precedent of theoretical framework to support”--------a reformed theory or a new theory is on the way, so “hypotheses can be formed for testing” based on the “reformed theory or a would be new theory”.

Yours,

Geng

Hi Geng,

In the absence of precedent theory, a new / reformed theory will be formed after your current research model / conceptual framework (including its hypotheses) being empirically tested and showing evidence to support your new / reformed theory.

Regards,

Fung

Hi Fung,

In my personal view, hypotheses should be conducted by “ideas” and ideas are parts of the “reformed theory or would be new theory”.

Hypotheses are other names for “ideas”.

Regards,

Geng

Gueng

Sorry, no. A hypothesis is quite different concept from idea, "hypothesis" intention and extension being much narrower and specific, and therefore different.

Dear Mr. Dragan Pavlovic

So, we may come back to the definitions and cognitions of “ideas” and "hypothesis". Of cause “ideas” are not exactly the same as "hypothesis"--------they are two concepts.

What are the connections and differences between them?

Regards,

Geng

Dear Geng,

There are many sources on the Internet that offer long discussions on ideas or hypotheses. We should in our discussion try to go beyond those basic notions. OK, anyway. A hypothesis is a proposition (that may be true or false). An idea may be probably also a mental representation of a proposition, but also of anything, even of the concepts that are abstract, inexistent, of feelings, intentions, words, and often may have very complex mental contents.

Dear Mr. Dragan Pavlovic

I agree with you that "hypothesis" intention and extension are much narrower and specific than “ideas”. Are following conclusions logical?

1, all "hypothesis" belong to “ideas”, they are a kind of “ideas” though with different forms

2, “ideas” include all "hypothesis" but "hypothesis" do not include all “ideas”-------no bijections between them.

Regards,

Geng

Dear Geng,

Yes, I think that a scientific hypothesis is probably always an idea, a proposition, explanatory or one that states some facts which should be then verified and if proved to be the case, would constitute some form of a theory. Yet, hypotheses that we make all the time in our everyday life about the world do not have to be always well defined ideas but they are certainly propositions that are often non-“represented”, do not have in the background defined ideas.

We perform almost all the time acts that are based on our hypotheses about the world, but we do not formulate them as some clear hypotheses. We perform number of acts that look like simple reflexes: silence our alarm clock in the morning with a typical arm movement and exercise an action on the button of the clock that stops its ringing. We do things all the time that are based on our knowledge about our external world and on hypotheses about the effects of our movements on the external world and our body. We of course do not have behind some clear “ideas” at those moments.

When for example plying table tennis we see a ball coming to the edge of the table at a certain angle we may “predict” its irregular bouncing and try to get that ball at some not well defined place in the space – without really thinking. Yet if asked how we managed to get that ball we can remember that we knew that that particular edge of the table had a defect and that we predicted where the ball would be at the next instant! This is quite extraordinary, but it may be that we may have some very precise ideas also when we think that our reactions were only a simple reflex. I do not include here a simple spinal reflex that no doubt does not include immediate cortical representations and of course neither some “ideas”.

Similarly, and much more reduced example would be when a spider that, sitting in its corner, feels the vibrations of its net and sees a fly trying to free itself of a deadly trap, immediately runs in the middle of the net to finish its pray. The spider certainly does it as a reflex which has in fact a logical form of a hypothesis. I do not think that a spider had some idea at the same time, but the hypothesis, i.e.; the presence of a clear propositional “content” is obvious. There is of course a big difference between the two but when a reflex becomes a hypothesis, and a hypothesis, that is only a simple non represented proposition, becomes a complex hypothesis that has an idea as a part of its “structure” - is hard to say.

This is a hard subject and I would avoid developing a full treatise about this here.

Mr. Ouyang, your conclusions are logical and therefore correct, given the current level of our understanding and knowledge in this area. I would just like to add to this my bit. In my humble view a hypothesis is probably that idea (ideas) which exists in the form of a tentative proposition which needs to be tested and verified and then developed as a part of a proper theory. Whereas the theory would be a more developed set of connected ideas, which after having been tested and verified through experimentation are at a particular time viewed as correct and are possibly the most appropriate explanation of a phenomenon that one is trying to understand. Here i would like to also take the opportunity to mention an observation.

It is correct that most of our hypotheses are based on relevant or irrelevant theoretical frameworks and that is due to how the post language human mind has developed during the period of civilization. During the last 6000 years or so human beings have accumulated a vast and unending ocean of knowledge fund in the form of ideas, which have been chiseled, developed and applied in human societies. So today when we make hypotheses then most of the ideas we use in them are a part of some theory. In fact scientists like David Deutsch have gone so far as to suggest that "All observation is theory-laden". I agree with this in principle because our pre-existing ideas do and can influence our observations. But then i cannot make this a dogma that it has to always be so.

If one is inquiring into a phenomenon about which existing theories are not providing adequate and clear ideas which can provide a better and more real understanding and if one is aware of the influence of pre-existing ideas on one's observations (whether sensory or mental) then it could be possible to undertake the making of those hypotheses which are not constrained or guided by the pre-existing idea frameworks. Of course during hypothesizing one can use some relevant pre-existing ideas but these are only as an aid or tool and not for channelizing the process. Then one could make new hypotheses and then test and verify them in order to make a new theoretical framework which can enable one to move onto a higher and better plane of understanding that phenomenon. I think probably that is how new and novel understanding has arisen in science and in other areas of knowledge also. Hypothesis making undertaken by the intellect can and should be able to rise above given theoretical constraints, while trying to break new grounds in understanding a phenomenon. Otherwise it will remain largely constrained and chained by the existing idea frameworks and not be able to move towards serious autonomy. Which is what is needed in this present highly complex stage of human existence and doing.

I think one needs to understand more clearly and in more detail all these various classifications or distinctions; hypothesis, idea, theory, conjecture, conclusion, etc. How they are interconnected but also distinct, where they overlap and where they need to be separated. Where we sometimes use them interchangeably and probably create confusion. I guess we need to keep improving our understanding about how the mind works; the mechanics of its structure and the various functions, how it operates and how and why it emerged and then evolved to reach its current stage in human beings, and how it is a part of the larger journey of Nature's own evolution and so on. We need to know it much more than we know at present. That will introduce more clarity in our minds and remove confusions and misunderstandings about the existing knowledge generation process which the mind undertakes.

I think Mr. Pavlovic is right that reflexes also have some form of ideas behind them. Of course they are largely nonverbal ideas, which in my view exist in the form of mental and physical programmes in the genes, brain and mind. There are mental reflexes also when we think about something in the same manner again and again...those are mental reflexes and habits. Similarly you have simple reflexes at the physical level which are the activation of programmes in genes and the brain. So nonverbal ideas are encoded in our reflexes also. I will end here now as this must have become monotonous for all my colleagues.

A few readings that may be of interest to this thread:

(1)         Lieberson, S., & Horwich, J. (2008). Implication analysis: A pragmatic proposal for linking theory and data in the social sciences. Sociological Methodology, 38(1), 1-50. doi:10.1111/j.1467-9531.2008.00199.x

(2)         Singer, B. (2008). Comment: Implication analysis as abductive inference. Sociological Methodology, 38(1), 75-83. doi:10.1111/j.1467-9531.2008.00209.x

(3)         Greenwald, A.G., Pratkanis, A.R., Leippe, M.R., & Baumgardner, M.H. (1986). Under what conditions does theory obstruct research progress?  Psychological Review, 93, 216-229.

(4)         Wallach, L., & Wallach, M. A. (1994). Gergen versus the mainstream: Are hypotheses in social psychology subject to empirical test? Journal of Personality and Social Psychology, 67, 233-242.

(5)         Chow, S.L. (1992). Acceptance of a theory: Justification or rhetoric? Journal for the Theory of Social Behaviour, 22, 447-474.

Dear Alex Chris.

 Nowadays it is in the air the notion of "discovery driven science", specially in molecular biology, where the overwhelming data collections cannot be related with any previously known theory, so the role of theoretical thinking is to expand the rational ability to construct new conceptual horizons. Errol E. Harris points on the role of "preparedness of mind" and he is aware that humongous data collections cannot be the formal basis of new knwoledge, so the important role of theory is to enhance our perception through the rational exercise, scientific research is mainly an abstract exercise, the instrumental phase of knowledge, cannot go further without the main guide of theory.

  From my point of view the only fact to be able to propose or search a hypothesis, involves a natural movement towards the theoretical way of thinking, with all its scientific techniques and methods, no matters if it is within natural or social sciences.

  This book is not a new one, but it has proven to be of great value to form the scientific perception.

Errol E. Harris. Hypothesis and Perception: The Roots of Scientific Method. (Humanities Paperback Library) 

Have a look at the paper "Discovering Data on Religion", chapter 8 of the book "Exploring the Religious Life", by Rodney Stark, Baltimore, The Johns Hopkins University Press, 2004, pp. 162-183. It answers your question without much ado.

Brian, Stefan, Fernando

Thanks for the suggestions. But: Please do not give us homework to do.

This is ReserchGate, there are people on the site who are already IN the particular research in question. Please give an answer if you have one; or tell us what is the answer that may be correct - even if it is in the books that you suggested. Give its short version that will be understandable, give us a paragraph that answers the question. Please.

Gabriel

I hold that our hypothesis is already contained in our scientific method, more precisely in the way we grasp the results, register, and measure. We may of course project further of what we will observe and form hypotheses that contain deeper theoretical explanations. Those may be presently not measurable but should be measurable in principle.

"Can someone refer me to a research paper that emphasizes the need for theory driven hypotheses?" I did answer that question.

Fernando

Do you think this is the place for listing the references, even the references on religious philosophy?

Why did you vote me down? We vote down here only bad, unfriendly answers.

You can find good examinations of some of the issues surrounding the good and bad of hypotheses in:

Greenwald, A.G., Pratkanis, A.R., Leippe, M.R., & Baumgardner, M.H. (1986). Under what conditions does theory obstruct research progress?  Psychological Review, 93, 216-229.

Wallach, L., & Wallach, M. A. (1994). Gergen versus the mainstream: Are hypotheses in social psychology subject to empirical test? Journal of Personality and Social Psychology, 67, 233-242.

Chow, S.L. (1992). Acceptance of a theory: Justification or rhetoric? Journal for the Theory of Social Behaviour, 22, 447-474.

I wrote a lengthy essay but unfortunately it disappeared. Sorry, Alex.

If you are using a quantitative methodology, you must have at least one hypothesis.  Without a theoretical or conceptual framework, it is just a collection of interesting facts.  Jan Vinita White, PhD

A hypothesis is nothing but a synthetic statement not yet verified. A syntehtic statement is made on the base of a fact or a few facts verified by the researcher adding to it necessity and universality. When the statement is based on a fact not verified or invented that statement is a not synthetic but a hypothesis thus needing verification. if a statement once stated is repeatedly verified it is true. It constitutes science. Otherwise it is false. Statements usually belong to a branch of science. They all are based on a theory of knowledge, that is, they are possible because of the human faculty of knowing. You can see Kant (synthetic a priori statements) or Ortega y Gasset (synthetic statements) in any dictionary of phylosophy.

Alex, you might do a search on hypothetico-deductive method. See

https://www.britannica.com/science/hypothetico-deductive-method