Social Research Glossary


A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Home


Citation reference: Harvey, L., 2012-19, Social Research Glossary, Quality Research International,

This is a dynamic glossary and the author would welcome any e-mail suggestions for additions or amendments. Page updated 23 January, 2019 , © Lee Harvey 2012–2019.


A fast-paced novel of conjecture and surprises



core definition

Probablism is the sceptical view that no definite knowledge can be obtained and, therefore, opinions and actions should be guided by probability.

explanatory context

Probabalism is a modification of inductivism that asserts knowledge is the result of probabalistic inference. (It was proposed by a group of Cambridge philosophers.)

Probabalism accepts the anti-inductivist position that theories are unprovable on the basis of empirical data (i.e. that inductive 'proof' cannot be generalised).

However, proababilism accepted that some theories are more probable than others in the light of the available evidence. Scientific honesty is then reduced to declaring only highly probable theories and the justificationist dictum of scientific honesty appertaining to proven propositions is therefore discarded.

However, this watered down version of inductivism does not overcome the problem of inductive inference as there is no way that any finite number of observations makes a theoretical proposition more probable given the infinite implication of such a generalisation. The probability of such a proposition being true is still zero however many observations appear to support it. The proposition may, of course be intuitively more true, but logically this is irrelevant and certainly does not provide an objective support to the inductivist principle.

An alternative probabilistic manouevre is to frame science as about particulars rather than universals. To make it a predictor of unique events. For example, to predcict the probability of the sun rising tomorrow rather than always. The probability is thus a function of available data and calculable. But, all that it amounts to is a particular production. It does not constitute scientific knowledge as such, for scientific knowledge is entwined with universalistic comprehension, to which particular predictions offer nothing.

Furthermore, any prediction is credible in relation to a theory that suggests what evidence the probability should relate to. A prediction devoid of theory is useless in the development of knowledge. For example, it might be possible to predict the winner of a horse race simply by computing the probability of each horse winning on the basis of their wins in other races. If, however, the prediction is based on a more sophisticated theory, which takes into account not only the form of the horses but the opposition the horses faced in other races, the breeding of the horses, the length of the race, the jockey, the training, the weight allowance, the course and condition of the ground, and so on, then the prediction will be more credible because all relevent factors will have been accounted for. The relevence of these factors is, of course, theoretically determined.


Without labouring the point further, the attempt to save inductivism through the weakening of its principles embodied in probabalism fails as all theories are not equally unprovable but are just equally improbable.

analytical review

associated issues


related areas

See also



copyright Lee Harvey 2012–2019


A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Home