The sound of a bell and the spoken word bell are different things. Carl Wernicke used that distinction in his 1874 book on language difficulties. He described a concept of a bell as linked memories of seeing, touching and hearing it. The word used to name that object was a conventional sign, not another physical feature of the bell (Wernicke 1874, digitized copy, p. 36).
This example helps explain why his model contained more than two isolated places, one for speaking and one for understanding. It also needed connections between remembered sounds, movements and concepts.
A book and a career
Wernicke lived from 1848 to 1905. Born in Tarnowitz in Upper Silesia, he studied medicine at Breslau. Michael Sachs’s scholarly biography dates his medical doctorate to 1870 and his move to the Charité in Berlin to 1876. He became an associate professor at Breslau in 1885 (Neue Deutsche Biographie).
His Der aphasische Symptomencomplex appeared in Breslau in 1874, published by Max Cohn & Weigert. Its subtitle describes a psychological study grounded in anatomy (Wellcome catalogue and digitized copy).
Later, Wernicke and Theodor Ziehen founded Monatsschrift für Psychiatrie und Neurologie in 1897 (publisher’s journal history). The monograph and the journal were different kinds of work.
Parts that work together
Wernicke did not try to place every complicated mental activity in a separate patch of brain. Early in the book, he assigned simpler functions to particular regions, while describing thinking and the linking of impressions as activities involving connecting fibers. He credited Theodor Meynert’s association systems in this explanation (p. 4).
For spoken language, he proposed that the superior temporal region (the upper part of the brain’s side, or temporal, lobe) held remembered word sounds, while the inferior frontal region (the lower part of the front of the brain) held representations of the movements used to speak. He placed a connecting route through the insular region, a part of the brain lying beneath the surrounding surface folds (pp. 18–19).
These were components of his proposed explanation. Naming them does not establish that the brain contains literal pictures of sounds or a single store of speaking movements.
Hearing, naming and repeating
Wernicke’s diagrams separated several tasks. Hearing a word could activate its remembered sound and connect it with a concept. Naming an object could begin with its sensory features and lead toward the movements needed to say its name. Repetition connected a heard word with its spoken form (pp. 19–23, 36).
He also gave remembered sounds a checking role during speaking. In his account, they helped constrain which movement representation was selected. If that influence was lost, he expected word substitutions as well as difficulties understanding speech (p. 23).
Documented as a historical model: these are claims about the explanation Wernicke proposed. They are not a description of every process established by present-day research.
Why a broken connection mattered
The model allowed more than damage to one center or another. Wernicke also considered damage to the fibers connecting word sounds with speaking movements while the representations themselves remained available (pp. 26–27).
He predicted a pattern in which understanding could remain while word selection during speech was disturbed. This was part of his account of conduction aphasia. The important historical move was to ask what a damaged connection could explain, as well as what a damaged region could explain.
That argument should not be compressed into “Broca found speaking, and Wernicke found understanding.” Such a summary leaves out the pathways that made his model work.
Observations had limits
In the book’s third part, Wernicke contrasted motor and sensory forms of aphasia and presented case histories (pp. 38–39). He criticized earlier reports for incomplete descriptions and uncertain anatomical evidence. Yet he also identified a weakness in his own records: the notes on Susanne Rother lacked the detail and precision he wanted because his analysis of the problem had not yet been sufficiently developed (p. 43).
Near the end, he acknowledged that his views were not wholly new and that some anatomical questions remained unsettled (pp. 67–68). These admissions belong alongside the diagrams. A model can organize observations while still resting on incomplete evidence.
Later diagrams were revisions
Ludwig Lichtheim published his own diagram in 1885. Much later, Norman Geschwind revived this tradition, but his version differed from Wernicke’s original. Ardi Roelofs’s 2024 comparison traces these changes and cautions against treating the familiar later model as an unchanged copy of the 1874 book (historical analysis).
How we know
What survives: The original monograph, its diagrams and case histories, later publications and biographical records.
Whose voices are missing: Patients’ experiences are filtered through the physician’s questions and descriptions.
What is uncertain: The historical diagrams combine observations with proposed mechanisms and incompletely established anatomy.
Then vs. now
Wernicke’s contribution included relationships between processes, not simply labels on a brain map. Keeping the original proposal separate from later revisions lets us see how the explanation changed.