Was the QWERTY keyboard created because people could type too fast with an ABCDE keyboard?
No. The Sholes layout was created because, while the alphabetical layouts that came before it were easy to learn and show off to potential buyers, they were rubbish in actual use. He reasoned that an efficient keyboard needed to put the letters most often used in English in positions where they were easiest to strike, and balanced between the hands to prevent jamming from occuring when adjacent keys were struck in rapid succession. Using the emerging research on letter frequency and a long series of experiments, he created a keyboard that did just that — allowing typists to speed up to previously unheard of velocity, as fast as their fingers and the mechanism could tolerate.
This was the first keyboard layout to be of actual use in practice, and was widely used by clerks transcribing telegraph messages — for whose convenience, a few small tweaks were made.
This created a market for the modern typewriter, and the expansion of business and government record keeping caused it to bloom in the early 20th century. The mechanism went through a series of further revisions, mostly by Remington and other companies involved in the early market, to further speed up typing by reducing mechanical limitations (which persisted right up until IBM’s invention of the electric typewriter with its ball-shaped strike head).
Remington further corrupted the layout by making a few random changes intended to thwart existing patents, but the point was moot. Remington cornered the market, and their version of Sholes’ layout became the defacto standard.
And naturally, it isn’t optimal. However, careful studies by the US Navy and later by several universities have found that the Sholes, or QWERTY layout is sufficiently close to optimal that if another layout (specifically Dvorak) truly is superior, it is by too narrow a margin to measure (by typing speed, at least). This is confirmed by the fact that world’s records for typing speed have consistently bounced back and forth between the Sholes and other layouts. If the perceived suboptimality of its design really mattered, it would have been left in the dust long ago.
These studies have found that on any modern keyboard, typing speed is determined by effort and practice, with the layout having no measurable effect. So if you want to speed up typists, your time and money is better spent on letting trained typists practice on QWERTY than on retraining them on anything else. Furthermore, studies in the 1970s and 80s, still during the era of the typing pool, found that in practice, professional typists seldom exceed 35 words per minute, whatever their peak performance on typing speed tests.
Now, the Dvorak devotees and the “cult of the superior keyboard” will now chime in that even if all this is true, their favorite keyboard is more comfortable. Good, then use it. No one is stopping you. But the human hand is among the most variable biomechanical devices in all of nature. What works for you doesn’t work for everyone, so experiment and find your sweet spot.
Actual science shows, however, that just as speed is not determined by keyboard layout, ergonomic performance is more impacted by posture, habit, and workstation design. Fundamentally, any flat keyboard will do a poor job of fitting human anatomy, but any contoured keyboard will be harder to use.
As usual, life is a bag of trade-offs.