QUANTITATIVE ESTIMATES OF SENSORY EVENTS 305 



assumption of the constancy of natural laws. Any such assumption belongs 

 to metaphysics and has no place in physical science, in which no statement 

 about the behaviour of the universe can be made except on the basis of 

 experimentally determined facts ; and no facts about the relations between 

 phenomena and temperature can be determined until a metrologically 

 sound method of measuring temperature is available. We must therefore 

 be satisfied that our method of measurement is sound without reference to 

 any natural laws involving temperature. The point is that the constancy 

 of the law defining our scale does not require confirmation. It is not an 

 assumption, which may or may not be true, it is a postulate forming part of 

 the conventional framework of physical measurement. The postulated law 

 is necessarily always true for the simple reason that it serves the purpose of 

 defining temperature as ' the thing for which this law is true.' There is no 

 criterion of the magnitude of a temperature (nor of any B magnitude) other 

 than the law by which we choose to define it. It would therefore be 

 meaningless to ask whether the temperature to which our scale assigns the 

 numeral 7i is in fact the same temperature at all times and places. 



Temperature only enters into the physicist's experience indirectly, in 

 virtue of its effect on the measurable properties of bodies. He is therefore 

 never concerned with it as a thing having existence in its own right which 

 may on occasion assert that existence in ways which are inconsistent with 

 his arbitrary definitions. He is perfectly free to define it by any postulated 

 relation to some one physical property, and so long as he keeps to this 

 definition and makes his measurements in accordance with it no contra- 

 diction between the results of measurement and any other kind of experience 

 can ever arise. 



Psycho-physical Measurements. 



With the foregoing introduction in mind, the application of measurement 

 to psycho-physical problems may now be considered. The title of this 

 Committee is wide enough and vague enough to include many matters 

 which are not in dispute, and some misapprehensiorthas been evident during 

 the deliberations of the Committee as to precisely what is in dispute. Those 

 who have contended that methods hitherto supposed by some to lead to 

 quantitative scales of sensation intensity are invalid have been supposed to 

 contend that no quantitative experiments bearing on ' sensory events ' are 

 possible. Such a contention would of course be absurd, and I will first 

 consider some important types of psycho-physical measurement which can 

 be, and have been, effected. These may briefly be classed as measurements 

 of relative stimulus efficiencies. An important and well-known example of 

 this kind of measurement is found in the so-called relative visibility function 

 of the eye, which tells us the relative visual efficiency of unit intensity of 

 monochromatic radiant energy of different wavelengths. Briefly, these 

 measurements consist in determining the relative amounts, measured in 

 physical units, of two samples of monochromatic radiation of different 

 wavelengths which are required to make the two sides of a photometric 

 field appear equal with respect to the attribute of brightness. By repeating 

 this for different pairs of stimuli we obtain measures of the relative physical 

 intensities of stimuli of various wavelengths throughout the spectrum which 

 are required to produce equal intensities of the sensation of brightness. 

 The relative efficiencies of radiation of different wavelengths for exciting 

 the sensation of brightness are inversely proportional to the amounts 

 required to produce the same brightness, and it is these relative efficiencies 

 which are usually tabulated or plotted as the ' relative visibility function ' of 

 the eye. 



