Diagramming: history of the visualization of grammar in the 19th century
« previous post | next post »
Aside from etymology, one of my favorite language study activities before college was diagramming sentences. Consequently, I was delighted to be reminded of those good old days by this new (June 19, 2024) article in The Public Domain Review: "American Grammar: Diagraming Sentences in the 19th Century". This is a magisterial collection of crisply photographed archival works that you can flip through page by page to study at your leisure.
The works collected are the following:
James Brown, The American Grammar (Philadelphia, PA: Clark and Raser, 1831).
Frederick A. P. Barnard, Analytic Grammar; with Symbolic Illustration (New York: E. French, 1836.
Oliver B. Peirce, The Grammar of the English Language (New York: Robinson and Franklin, 1839).
Solomon Barrett, The Principles of Grammar (Cambridge, MA: Metcalf and Co., 1857).
Charles Gauss and B. T. Hodge, A Comprehensive English Grammar (Baltimore, MD: Pan Publication Co., 1890)
Stephen Watkins Clark, A Practical Grammar (New York: A. S. Barnes and Co., 1847).
Alonzo Reed and Brainerd Kellogg, Higher Lessons in English (New York: Clark and Maynard, 1880).
When I was in high school, we were still doing diagramming like that in Reed and Kellogg (1880), and I loved (almost) every minute of it, although sometimes it was vexatiously challenging to make everything fit in neatly and rigorously.
The text, by Hunter Dukes, is both entertaining and edifying, although the introductory quotation is rather impenetrable:
“Once you really know how to diagram a sentence really know it, you know practically all you have to know about English grammar”, Gertrude Stein once claimed. “I really do not know that anything has ever been more exciting than diagramming sentences. . . . I like the feeling the everlasting feeling of sentences as they diagram themselves.”
In my estimation, Stein's sentences would have benefited from the addition of a few commas.
Dukes follows thus:
While one student’s lexical excitement is surely another’s slow death by gerund, Stein cuts to the heart of the grammatical pull. Is grammar prescriptive and conventional, something one learns to impose on language through trial and error? Or do sentences, in a sense, diagram themselves, revealing an innate logic and latent structure in language and the mind? More than a century before Noam Chomsky popularized the idea of a universal grammar, linguists in the United States began diagramming sentences in an attempt to visualize the complex structure — of seemingly divine origins — at their mother tongue’s core.
The collection then proceeds, one book at a time. It is interesting to observe how the explications and illustrations become increasingly clear and sophisticated through the years and decades.
Some highlights:
The history of diagramming sentences in the United States begins with James Brown’s American Grammar (1831). “Language is an emanation from God”, he writes. “As a gift, it claims our servitude; as a science, it demands our highest attention.” … It was in American Grammar that Brown debuted construing as a method for parsing sentences using a system of square and round brackets to isolate major and minor sections. Major sections are “mechanically independent”; minor sections are “mechanically dependent”. Brown called this form of analysis close reading, but construing was only one half of the system. “As construing is a critical examination of the constructive relation between the sections of a sentence, so scanning is a critical investigation of the constructive relation between the words of a section.” Scanning involves ranking minor sections in ascending numerical order based on their relational distance from the major section. Playing a kind of grammarian god, Brown uses John 1 to demonstrate how his system can cleave sentential flesh. (In the beginning) [was the word] (and the word was) (with God) (and the word was God).
Barnard's presentation is so highly logographic and isolating that it almost makes me feel that he was influenced by common misunderstandings of the the Sinographic script, which puts him quite at odds with the formidable (pronounce it in French, please) lawyer and linguist, Peter Stephen Du Ponceau (born Pierre-Étienne du Ponceau, June 3, 1760 – April 1, 1844), whose revolutionary A Dissertation on the Nature and Character of the Chinese Writing System, published in Philadelphia by the American Philosophical Society in 1838 (almost the same time as Barnard's tome), focused on the sounds and words of the Chinese writing system, not on its alleged universal, ideographic nature.
Whereas Brown sought to reform an educational system plagued by “simplifiers”, “plagiarists”, and “new modellers”, Frederick A. P. Barnard’s Analytic Grammar; with Symbolic Illustration (1836) reported from the classroom on syntactic techniques “advantageously used in the instruction of the deaf and dumb”. His system for diagramming sentences is mildly pictographic, based off of six principal symbols for marking substantive, attributive, assertional, influential, connective, and temporal words and phrases. Each symbol, in turn, can accommodate a host of diacritical marks to further specify its function. In the sentence “A man goes into a house”, for example, “man” is marked with a vertical line, signifying the noun’s substantive property. Two feet are added to the line since “man” is the subject of this sentence, “the supporter of what follows”. Because the word is in the nominative case, it also receives a diagonal, accent aigu–like appendage, pointing the action forward. “Goes into” necessitates a confluence of connected marks that resemble the Eye of Horus. First, we start with a horizontal line, the attributive verb. Since it contains an intransitive assertion, it receives a v-shaped hat, whose right arm curls in on itself, signifying “the attribute exists merely in the agent himself, without regard to any outward object”. Interlocked to this arm is a spiral-like symbol that accounts for the preposition “into”, “a connecting link”. “House”, in turn, looks a lot like “man” — built on a substantive vertical line — but with a grave accent instead of acute, symbolizing the objective case: receiving the action thrown forward by the nominative subject. Curiously, whereas Brown turns toward scripture for his corpus, Barnard’s examples frequently express physical violence or categorical division, mirroring the two-fold sense of “articulation” that his system embodies: both a means of expression and a form of dissection at the joints. “The victor exceedingly rejoices in his conquest”; “He is about to tear a book”; “Negroes are habitant in Africa”.
Goodness gracious! Barnard's "mildly pictographic" system even has six principal symbols for marking different types of words and phrases. This reminds me of the liùshū 六書 ("six writings", i.e., six types of character composition) of the Chinese lexicographer Xu Shen (c. 58–c. 148 AD): pictographic, indicatives, ideographs, phonetic compounds, mutual explanatory, and phonetic loans (source). Of course, Barnard's six principal symbols for types of words and phrases and Xu Shen's six types of character composition are quite different in their referents, but so uncannily similar in their application it's possible that, if Barnard had looked into what was known about the construction of the Chinese writing system at his time, he might have been vaguely inspired by some of the basic concepts of Xu Shen.
Three years later, Oliver B. Peirce’s The Grammar of the English Language (1839) came on the scene with a fervent rhetoric that reads as hyperbolic even among hot-headed grammarians. “On this imperishable foundation — this rock of eternal endurance — I rear my superstructure, the edifice of scientific truth, the temple of Grammatical consistency.” Far less systematic than Barnard, Peirce arranged his sentences with a chain-link structure: assertives and relatives (verbs and prepositions) connect larger subject and object circles whose articles are attached to these nouns like keys on a ring. The visual conceit is apparent and didactic: appendant clauses are literally appended one onto the next; if one link grammatically falters, the whole chain of meaning becomes undone.
Such overwhelming confidence in the truth value of grammar is breathtaking!
God (the trunk of der Grammatikbaum) hath spoken.
Before “syntactic trees” became common parlance for linguists, Solomon Barrett’s Principles of Grammar (1845) used a similar metaphor and added bark. The frontispiece displays Hebrews 1 as an old-growth hardwood: “God” is the trunk, the predicates “who spake” and “hath spoken” form solid boughs, while prepositional phrases are figured as finer twigs, pruned of all foliage. Ranging beyond his peers into Greek, Latin, French, Spanish, and German grammar, Barrett’s arboreal figures suit his interest in language’s branching connections, the root-like structures of etymology and inheritance.
The central role of the tree metaphor is noteworthy in many of these 19th century presentations on grammar visualization.
It is not until Stephen Watkins Clark’s 1847 work, A Practical Grammar, however, that we find a system that strongly resembles sentence diagramming in its modern-day — though quickly fading — guise. Combining the divisional schema of Brown and Barnard with the visual style of Peirce’s scalar links, Clark’s method uses word balloons that resemble, in the words of Kitty Burns Florey, “elaborate systems of propane storage tanks — or possibly invading hordes of Goodyear blimps”. There are twelve general rules and scores of definitions that resemble mathematical proofs. A sentence’s principal elements occupy the highest row. Subject, predicate, object — there is a fixed order of operations. Adjuncts are placed below the words they limit or modify, conjunctions between the terms they join, and pronouns dangle from their antecedents by umbilical cords. Clark’s enduring innovation was attributing properties to “offices” rather than individual words — offices that could be occupied by words, phrases, or even entire sentences. Grammar thus becomes a system of scalable relations rather than a paint-by-numbers tool for classifying parts of speech. “Major grounding ideas still present in modern IC [immediate constituent] analyses and PSG [phrase structure grammar] were already present in Clark’s syntactic conception”, writes Nicolas Mazziotta. Comparing grammar to “the foundation of a building”, Clark gave his students a toolbox for dismantling faulty foundations and properly assembling sentential edifices of their own.
Now we come to the "modern" diagramming with which many of us are familiar.
Like the twisted balloon animals that they resemble, Clark’s annotations floated into his contemporaries’ linguistic consciousness. In Higher Lessons in English (1877), Alonzo Reed and Brainerd Kellog deflated his bubbles into lines but preserved many of their predecessor’s innovations. Returning to Enlightenment preoccupations, Reed and Kellog begin their treatise with a discussion of the natural language that “we never learned from a grammar or a book of any kind”: “the language of cries, laughter, and tones . . . the language of gestures by the hand, and postures by the body”. While this form of human signification is purely innate, they claim, spoken language (or “Word language”) must be governed by a grammatical “science which teaches the forms, uses, and relations of the words of the English language”. Their system and subsequent companion volumes were so popular that, for a time, the pair’s books sold a half-million copies per year. As Richard Hudson notes, the Reed-Kellog system for diagramming sentences is still taught in American schools.
At that point, the text comes to a close. It is followed by a gallery of fourteen sentence diagrams gathered from the works discussed.
Visualization is a key component of Buddhist thought and practice, e.g., here. "Visual metaphors, visionary literature, and visualization practices are pervasive in Buddhist traditions. Vision and seeing are dominant metaphors for knowledge, awakening, and insight." Perhaps Western grammarians were also inspired by becoming familiar with this aspect of Eastern thought as well.
An added treat are links to eight visually oriented "related collections", several of which I find highly attractive:
Mnemonic Alphabet of Jacobus Publicius (1482)
A Dictionary of Victorian Slang (1909)
Growing Things: A Film Lesson in “Nature Study” (1928)
Arthur Wesley Dow’s Floating World: Composition (1905 edition)
Punctuation Personified (1824)
The History of Ink: Including its Etymology, Chemistry, and Bibliography (1860)
Aspirated Aspirations: Alfred Leach’s The Letter H (1880)
Humanity 101: The Syllabus of Frankenstein's Monster
Until I set about preparing this post, I had never heard of The Public Domain Review. Now I must say that it has found a soft spot in my heart.
Selected readings
- "Sentence length and syntactic complexity" (3/29/22)
- "Sentence diagramming" (1/1/14)
- "Personal and intellectual history of sentence diagrams" (10/14/04)
- "Diagramming Sentences" (4/14/13)
- "Putting grammar back in grammar schools: A modest proposal" (12/25/13)
- "School grammar, Round two" (12/30/12)
- "Diagrammatic excitement" (3/27/12)
- "Defiant diagramming" (10/5/08)
- "Not just Oxford commas" (6/25/24)
- "Peter Stephen Du Ponceau and Vietnamese dictionaries" (2/7/22)
[Thanks to Geoff Wade]
David L said,
June 28, 2024 @ 12:24 pm
Gertrude Stein was notoriously a comma-hater.