With the help of 26 other annotators, I tried to understand more about Doug Engelbart’s visionary text “Augmenting Human Intellect“, and especially to understand why some visions about amplification have not come true, despite his plausible argumentation.
He hoped that the human intellect would not only be augmented by a smart computer “clerk” (such that their combined capacity would increase), but that the intellect would indeed be amplified through using this clerk. He makes this plausible by extending the Whorf hypothesis. Whorf says that our thinking is affected by using the ‘tool’ of language. And now Engelbart extends this to the synergistic system of “H-LAM/T” “(Human using Language, Artifacts, Methodology, in which he is Trained)”: The combined ‘tools’ L, A, M, and T together will then similarly affect, and amplify, H’s intellect.
Why didn’t this work? What went wrong?
I think if we carefully read how Engelbart imagined his clerk, in contrast to how we see our modern computers, we might get a hunch.
My observation was in particular, that he did not describe such a fixedly defined way of interaction with, and separation from, the clerk as the modern ‘interface’ that we are used to. Rather, the user seemed to continually readjust this division of labor. The user “would find it very natural to develop further techniques on their own“, and he would offload and externalize a very internal part of their thinking, and would entrust it temporarily to the clerk.
This externalizing of thoughts, may seem difficult to understand if all we can imagine to be externalized is words and sentences, just as they are uttered or scribbled — as if thoughts all consisted of words and sentences. Piaget asked children what they use for thinking, and they responded they think with their mouth. And if we use our computer just like a better typewriter, it’s not much different.
In this case, even Engelbart’s clerk cannot amplify us any more. And the extended Whorf hypothesis will not work, but we will stick to the basic Whorf hypothesis — which also lends itself to an explanation why we equate language with thinking: language has so much amplified our thinking that our raw, unverbalized concepts now seem inferior. But it’s them that Engelbart’s vision builds upon. The concept structures lead to symbol structures, and these, in turn, can be externally manipulated.
(For more details, you may need to refresh the hypothes.is browser view. In many details, I found confirmation for the design of my tool, in particular the single notecards, the split screen, the size ratio, and in many places, the big emphasis on rearranging.)