Alfred Binet, answered from the texts and cited to the page.
When others set out to measure the mind, they measured what was easiest to measure: the keenness of the senses, the speed of reaction, the strength of the grip — supposing that the fine intellect must be served by fine instruments.1 But when these sensory tests were tried against the one thing we wished to predict — the child's power to think, to learn, to succeed in school — they predicted nothing.2
I took the opposite road. If we wish to measure intelligence, we must set the mind its own proper work: to judge, to comprehend, to reason, to find the point of a story, to see the absurdity in a picture, to define, to compare, to complete a thought. These complex operations are what intelligence is; the elementary sensations are its raw material, not its measure.
My scale is a graded ladder of such judgments, and it works because it asks of the child the very kind of thing that thinking is. To measure the higher faculties you must engage the higher faculties, not their sensory servants.345
Laura Bridgman, Helen Keller and their fellow-unfortunates were blind as well as deaf, but this did not prevent them from being very intelligent. Certainly this is demonstrative proof that the total or even partial integrity of the senses does not form a mental factor equal to judgment. We may measure the acuteness of the sensibiUty of subjects; nothing could be easier.The Development of Intelligence in Children, pp. 43–44
We therefore propose to make again the immediate repetition of the 8 sentences, by a general repetition. It is probable that the work which we announce will be completed by the time ^he present article is printed. TESTS OF SENSORIAL INTELLIGENCE These are made independently of the development of language, of the abstract idea, and have an extra-scholastic character. They are important from many points of view.The Development of Intelligence in Children, pp. 112–113
LAST REVISION 1911 4. How can teachers, by their own means, estimate the intelligence of a child? 5. What differences exist in the intelligence of children belonging to different social conditions? 6. What are the differences between our method and the methods with tests not arranged in a hierarchy? 7. Review of several recent works which have criticized our method.The Development of Intelligence in Children, pp. 275–276
Should we put to the test all of his psychological processes? A slight reflection has shown us that this would indeed be wasted time. It seems to us that in intelligence there is a fundamental faculty, the alteration or the lack of which, is of the utmost importance for practical life.The Development of Intelligence in Children, p. 43
During this maximum of twenty minutes, it must be turned and turned about in every sense, and at least ten tests must be executed, so that not more than about two minutes can be given to each. In spite of their interest, we were obliged to proscribe long exercises. For example, it would be very instructive to know how a subject learns by heart a series of sentences.The Development of Intelligence in Children, p. 42
This is one of Binet's most secure and consequential methodological contributions, now essentially uncontested in its core. The Galton- Cattell program of measuring intelligence through sensory acuity and reaction time failed empirically — the sensory tests did not predict academic or life achievement — while Binet's complex-judgment tasks did, and every subsequent intelligence and ability test follows his model of sampling higher-order cognitive performance rather than elementary sensation. The live modern debate is not whether to measure complex judgment (Binet won that) but exactly what construct the judgment tasks capture and how unitary it is — the g-factor and multiple-intelligences discussions that came after him.