ult from _improved methods_ of work. That was probably the case
with the telegrapher.

[Footnote: A plateau of this sort is present in the learning curve
for mastery of a puzzle, given on p. 316.]

The telegrapher acquires skill by improving his methods, rather than
by simply speeding up. He acquires methods that he didn't dream of at
first. At the start, he must learn the alphabet of dots and dashes.
This means, for purposes of sending, that he must learn the little
rhythmical pattern of finger movements that stands for each letter;
and, for purposes of receiving, that he must learn the rhythmical
{323} pattern of clicks from the sounder that stands for a letter.
When he has learned the alphabet, he is able to send and receive
slowly. In sending, he spells out the words, writing each letter as a
separate act. In receiving, at this early stage, he must pick out each
separate letter from the continuous series of clicks that he hears
from the sounder. By degrees, the letters become so familiar that he
goes through this spelling process easily; and, doing now so much
better than at the outset, he supposes he has learned the trade, in
its elements, and needs only to put on more speed.

But not at all! He has acquired but a small part of the necessary
stock-in-trade of the telegrapher. He has his "letter habits", but
knows nothing as yet of "word habits". These gradually come to him as
he continues his practice. He comes to know words as units, motor
units for sending purposes, auditory units for receiving. The
rhythmical pattern of the whole word becomes a familiar unit. Short,
much used words are first dealt with as units, then more and more
words, till he has a large vocabulary of word habits. A word that has
become a habit need not be spelled out in sending, nor laboriously dug
out letter by letter in receiving; you simply think the word "train",
and your finger taps it out as a connected unit; or, in receiving, you
recognize the characteristic pattern of this whole series of clicks.
When the telegrapher has reached this word habit stage, he finds the
new method far superior, in both speed and sureness, to the letter
habit method which he formerly assumed to be the whole art of
telegraphy. He does not even stop with word habits, but acquires a
similar control over familiar phrases.

Higher Units and Overlapping

The acquisition of skill in telegraphy consists mostly in learning
these _higher units_ of reactio