The Rhetoric of Record-Keeping II:
Shorthand Writing

My goal here is to make evident the rhetorical force of shorthand writing.

Partly, I am responding to recent study claiming that record-keeping, as a genre, functions primarily as an instrument of social control. The narrow claim is that record-keeping imposes or preserves ideological hegemony by elevating some values and suppressing others in a given speech community (Schryer, l993).

In response, I suggest that historical study of record-keeping complicates that claim and shows its limitations. To illustrate, I will explore shorthand writing, a practice that evolved over centuries to meet a basic requirement of record-keeping, the practical yet profound demand to capture what people say as they say it. Shorthand mediates that demand and renders transient speech for the purposes of memory, including institutional memory.

As a medium, shorthand is not transparent. Like any other medium, it is material, historical, and human-made. Because it is symbolic code, it shapes what it communicates. As a trans-medium between speech and writing, it has an interesting relation to literacy.

As a cultural phenomenon in nineteenth-century England and America, shorthand became a profession.

For practitioners, shorthand demands high cognitive skills of attention, recognition, memory, conceptualization; social skills of collaboration; and high linguistic abilities.

My chief example of shorthand use is in governmental record-keeping, specifically for US Congressional hearings. Before going to the example, I will briefly chronicle shorthand's history.

The Long History of Shorthand

As soon as alphabets began, shorthands apparently began.

In 4th century (B.C.E.) Greece, symbol systems were in use for abstracting alphabetic letters into single strokes that represented a letter, which, in turn, stood for common words, suffixes, and prefixes in which that letter appeared. To describe these systems, historians commonly use the synonyms stenography (narrow writing), brachygraphy (short writing), and tachygraphy (swift writing). All functioned to enable writing compactly and/or fast. (Encyclopedia Britannica , 1909).

These inscription methods were used for business (recording deeds); education (recording lectures); law (recording judicial proceedings); religion (revising and annotating texts); and politics (recording orations).

Roman uses, drawing on Greek systems, began with the caesars and flourished, apparently, under the emperors (Gregg,l992). Shorthand was taught in Roman schools. Cicero's freed slave, Tiro, is generally described as authoring the first truly tachygraphic system, one that more or less enabled verbatim reporting of human speech, by codifying a set of forms--symbols that by their shape and position represented common words. Tiro also developed a shorthand labor force, slaves trained to use the symbols who were positioned around the forum in Rome to record debate in relays.

Tironian Notes
In Tironian shorthand, shown here as reproduced in Gregg (l992), symbols might be combined or invented on the spot. Recorders seem idiosyncratically to have done so for uncommon words. Supplements for special purposes such as a set of forms for scriptural terms in ecclesiastical shorthand were accretions to the basic Tironian system. French, African, Spanish and other national variations of Tironian symbols also developed wherever, first, the Roman empire's bureaucracy and, later, the Catholic church's bureaucracy, reached (Gregg, l992).

Under the Holy Roman Empire, shorthand was declared necromantic and diabolical (Gregg, l992). Its decline was reversed, apparently, in Renaissance, Reformation and Restoration-era England when its use spread for recording preachers' sermons and dialogue in stage plays. Later, during English political revolutions of the 17th and 18th century, with the encouragement of printers eager for news to sell and material to publish, speech in royal chambers, in courts, in parliament, in the streets and taverns were recorded for amusement or profit. Shorthand was widely self-taught as well as taught by individuals and by schools, particularly Jesuit schools. How-to textbooks proliferated. Examples include:

Alphabetic shorthands had several flaws. They were clumsy (no punctuation); error-prone (look-alike symbols for many consonants, no symbols for vowels); and they required massive memory (thousands of symbols to memorize, often with thousands more for special purposes). Nevertheless, according to anecdotal and scholarly histories of shorthand (there apparently is no definitive history) we owe to these systems, used either officially or unofficially, our knowledge of: Socrates' lectures; Cicero's speeches; Jesus of Nazareth's parables and sermons; Augustine's treatises; variants of dialogue in the productions of Shakespeare's plays; and speeches in the British House of Commons establishing freedom of the press (Gregg, l992).

Shorthand in the Congress

Thomas Jefferson, James Madison, and other politicians in British and early republic America used shorthands for diverse purposes. Using a personally developed shorthand, for example, Madison voluntarily recorded the l787 Constitutional Convention while actively participating in it. He commented later that the job nearly killed him. Nonetheless, he acknowledged, transcribing his notes of each day's deliberation enabled him to plan his own moves for the following day as promoter of the republican plan of government that was, ultimately, adopted. Madison's notes, published after his death, are the only record of actual discussions. Without his notes, we now would not know the deliberations that outlined American government, because the Convention met in secret. However, the more general significance of shorthand's history in American political life begins with the Congress.

When the first federal Congress convened in l789, no provision was made for recording debate. The Constitution mandated a journal to be kept by the House of Representatives and the Senate, but the journal was limited to a record of actions taken, not to include deliberation. The journal was intended for members' use, not publication. Why did the Congress not wish to record and publish its deliberation?

In other commentary on the rhetoric of nation-building, I discuss ideological, historical, and procedural reasons for early Congressional resistance to recording debate. It is a political decision by government, especially a new democratic government, to be or not to be open and recorded. The House of Representatives and the Senate decided this question differently. The House, elected by popular vote and constituted to represent the people directly and to initiate legislation, opened its sessions to observers and news reporters from the start. In contrast, the Senate, elected by state legislatures and constituted to represent indirectly and to react to House initiatives, met in secret.

Here, however, my focus is less on the politics than on the practices and conditions. The significant condition was that debate was oral; the significant practice was handwriting as the only method for recording oral debate.

When the House opened its doors, Thomas Lloyd, widely recognized as a skilled writer and teacher of shorthand, who was also a committed patriot entrepreneur, entered along with many other free-lance reporters. Unlike the summarizers of debate for newspapers, Lloyd intended to report the full deliberations in the Congressional Register, a publication which he founded and edited. He did publish the first two sessions. Lloyd's Register was the most complete contemporary record. Until recently, it was the primary source for historical study of the first Congress.

Lloyd's recording method was an alphabetic shorthand such as I have described, the only kind available. Pages from his shorthand notes show Lloyd's "taking" (as the verbatim speech record is called by stenographers) that he later transcribed, edited, and published. Lloyd's notes are shown here as reproduced in Bickford by courtesy of the Library of Congress (Bickford, l992, 469). Their doodles and sketches suggest effects of some of Lloyd's working conditions: hours of debate by diverse speakers (some of them uneducated, others of them lawyers); many of whom he did not know by sight (and had to attribute, for example, as "baldheaded gentleman" or "man in blue coat and wig"), who were talking about difficult issues using complicated protocols for discussion.

Lloyd recorded daily, often exhausted from processing the previous day's notes. His process included staying to compare his "taking" with the clerk of the House's longhand notes, then transcribing. A reasonable estimate of the time required for transcribing Lloyd's "taking" is three to four hours per hour of debate (Tinling, l961). Sometimes, he acknowledged, he dozed during the sessions after long nights of transcribing. Other times, contemporaries alleged, he was drunk. That, too, Lloyd acknowledged. House members complained about the accuracy of published debates. Despite these complaints, Lloyd's records were at the time and until recently considered the best of the lot. His grave in Philadelphia carries a marker placed there by a nineteenth-century shorthand society honoring him as the father of modern stenography.

The processing required to reproduce a readable legislative debate from shorthand notes is considerable. Memory, comprehension, interpretation--indeed, composition--are required. Lloyd's shorthand, like most in his day, used look-alike symbols for many consonants; had no vowels; no punctuation; few articles and connectives; and many idiosyncratic abbreviations. Marion Tinling, who has transcribed Lloyd, produced the following breakdown of his process.

Processing Thomas Lloyd's shorthand

  1. Translation from shorthand to alphabetical reading:
    Bland jn n lwring fl s mc f ncsty f rsing rvn and vry frt gt t b t tt b wn the mns gt t tk sm t sbvrt thmslvs consdr bfr w dpt

  2. Transcription with vowels and punctuation added:
    Bland: Join in lowering--Feel as much of necessity of raising revenue--and every effort ought to be to that--But when the means ought to take seem to subvert themselves, consider before we adopt--

  3. Edited version with missing words inserted:
    Bland: [I] joint [with the gentlemen] in [the idea of] lowering [the duties]. [I] feel as much [as any one the weight] of necessity of raising [the] revenue, and [I feel that] every effort ought to be [directed] to that [end]. But when the means [we]ought to take seem to subvert themselves, [we must] consider before we adopt [them].

  4. Lloyd's own version:
    Mr. Bland: I join with the gentlemen who are disposed to lower the duties, although I feel the necessity we are under of raising the revenue as much as any other gentelman possibly can, yet I think we ought to deliberate fully upon the means before we adopt them.

Tinling comments that 18th century shorthand records, especially for interactive speech such as debate, were, at best, a prompt to reconstruction. "In general, verbatim notes could be transcribed only by the person who wrote them--and often not even by him. The reporter, having heard the words, used his notes to jog his memory, and his report was about as good as his understanding plus his memory" (Tinling, 530).

The Senate, in addition to its constitutional rationale for secrecy, sometimes justified remaining unrecorded on grounds that handwriting, the only available method of recording, was unreliable and error-prone. However, the invention in l844 of the telegraph, which carried instant reports of the Congress (errors and all) wherever the wires went, necessitated an accurate, authorized record. Fortunately, better shorthand methods and professionally-trained recorders were available by then. Congress sanctioned their use. In l848, both the House and the Senate officially authorized a record of their activity, put professional recorders on their staffs, and began publishing the regular report that became the present Congressional Record.

Improved nineteenth-century shorthand methods were phonetic, not alphabetic. In phonetic systems, consonant sounds (not letters) were represented by straight lines angled in one of four directions (horizontal, vertical, slanted left or right) and stroked either light or heavy according to which consonant was represented, e.g. light for p, heavy for b. Lines were hooked, sometimes with a circle added, to indicate consonant combinations, e.g. spr. Vowels were denoted by dots positioned relative to the stroked line for consonants, either beside, above, or below it. The great advantage of phonetic systems was speed--words and phrases could be written with fewer strokes. For accuracy in recording human speech, speed is what matters. Apart from this improvement, the phonographic writer's task remained much the same as when other systems were used--lists of standard forms still required memorization, and writers were encouraged to adapt the system as needed.

The best known phonetic shorthand, called phonography, was developed in England by Isaac Pitman and in America by his brother Benn Pitman. The Pitmans popularized and professionalized shorthand. Their how-to books were priced to put self-training within the reach of everyone; journals written in phonography were started; societies were formed to promulgate its use; competitions were regularly held to raise the standards of practice. Phonography spread through mass adult training classes before and after the regular working day, and, in England, through legislatively mandated training for military and legal professionals. For journalists, too, shorthand was a basic requirement.

Pitman's system was used by House and Senate recorders until approximately l971, when they switched to electronic machines with keyboards programmed to produce shorthand symbols on tape transcribed directly by an electronic typewriter (now by computer). Electronic shorthands use symbols other than Pitman, which is now archaic.

Portraits of Congressional Reporters of Debate

To suggest the attention, recognition, memory, conceptualization, and collaboration required by shorthand writing, I'd like to conclude this brief and somewhat technical history of shorthand with portraits of practice. Mark Twain or Charles Dickens might be portrayed, as they reported the Congress and the Parliament, respectively, for contemporary newspapers before they became better known as literary artists. However, they were not professionals. To portray reporting in government, I will summarize the perspectives of two career professionals: William Mohr, Senate Reporter, and Charles Reynolds, Senior Reporter in the Senate, whom I interviewed. (Mohr, Reynolds, l992)

Mohr, who has retired, was the last Pitman-trained reporter in the Senate. He is now transcribing historical journals composed in Pitman's phonography such as those by Montgomery Meigs, who in l852 was supervising engineer for the United States Capitol. Later, Meigs was Quartermaster General of the Union Army and corresponded regularly with his son, a Union soldier, using Pitman's phonography.

Meigs wrote Pitman script this way:

A page of Meigs' Pitman shorthand. Courtesy of William Mohr

Mohr translates Meigs' shorthand into his own adapted and simplified reporting style this way:

A page of Mohr's Pitman shorthand. Courtesy of William Mohr

Then, Mohr transcribes his own notes into readable pages of journal by Meigs:

(A-544) May 4th. This morning was perhaps the busiest day I have ever had. A perfect crowd of persons, contractors and others, looking at profiles, asking for places, etc.

At the Post Office, we are getting down well into the cellars. but I find there is to be some difficulty in getting a supply of water. The pipe at the Patent Office is not large enough. The foundations for our scales (?) is made and the timbers are being framed. The old houses are all down. The oxen are employed in plowing up the earth in the bottom of the cellars. . .

Mr. Rogers called to see me. He wished to begin one of the panels for the door, and for this purpose he asked me if I could furnish him with a room in which to study. I gave him a note to French, the Commissioner of Public Buildings, to try to get him one of the committee rooms in the old building.

I believe I was at my office for 2 hours or more before I could find time for that much conversation with him. I sent to Janes and Beebe the drawing of the window tracing of cast iron and asked them for an estimate or bid for their execution. The door tracings are nearly ready. I have quite a pretty design for the gate house at the falls, a plain Grecian for the body of the building, about 30 x70, with arched roof, the middle a pendentive dome, the order (?) to be Grecian Doric.

Berry and Mohon came to try to persuade me to buy their granite for the Post Office. Mr. Emery came with the same purpose, to sell his. They are jealous of each other. I promised to be receiving bids from them, to visit both quarries. . .

A constable came to offer his services as a watchman at the Post Office.

Transcript. Courtesy of William Mohr

The Meigs journals transcribed by William Mohr are published by the Office of the Senate Historian and the Capitol Historical Society in Washington, DC. They include information about the construction of the Capitol dome and Washington buildings. They report daily life in civil-war Washington, as well as Meigs' previously unpublished account of Lincoln's assassination.

In addition to Meigs' journals, Mohr has transcribed notes by women reporters at Eleanor Roosevelt's press conferences. These are original records of her interaction with women reporters to whom she talked directly. Mrs. Roosevelt admitted only women reporters to her press conferences. Women reporters were excluded from President Franklin Roosevelt's conferences.

In my interview with him, Mohr emphasized that interpretation is required in order to transcribe another person's shorthand notes. You are, in effect, reconstructing the recorder's understanding of a speaker's meaning, he said. Moreover, he added, you must interpret in order to record live speech, too. "You can't hear what you can't understand," he remarked. He described the self-education in medical practices and accident legalities that he undertook in order to report a Workmen's Compensation Court before he came to the Senate. (Mohr, l992)

Charles Reynolds, who until his death in l993 supervised daily production of the Congressional Record, began Senate reporting on the floor and in hearings in l948. Significant records he delivered as reporter include the McCarthy hearings on un-American activities in the Army and the McClellan hearings on organized crime in labor unions.

In my interview with him in l992, Reynolds emphasized the velocity of attention (my phrase) required of a legislative reporter. Unlike court reporters, recorders of legislative debate cannot stop the action to ask for repetition if they miss something. Whether listening to people speaking at the normal rate (approximately 150 words per minute), or at faster rates, or with interruptions, Reynolds said, the demand is the same. "You have to follow the 95-5 rule. Get 95 % the first time, because you can only confirm or correct 5 % later." Regarding the amount and kind of corrections a recorder might make, Reynolds followed an implicit standard called "Senate English" as it was explicitly taught to him by his mentors in Senate recording. The standard advises "Don't write what they said; write what they should have said (intended to say)." In my interview with him, Reynolds added that sworn testimony in a hearing, or testimony under oath, is always taken verbatim and is never corrected.

Reynolds described his personal method of preparation as applying the "45 second education rule," whereby he arrived early in the legislative chamber or hearing room to scan maps and charts posted on walls and repidly look through documents lying on tables. The second step was to listen with more than normal attention to the first speaker, in order to learn vocabulary and idiom within a subject area.

As to the physical stamina demanded of a recorder, Reynolds referred to a "50-50" requirement. "A Senate reporter's body must be 50% flesh and bone, 50% rubber band, so it can snap back again and again," he said. For example, Reynolds once "took" 37 hours of hearings at a stretch, with brief breaks to eat or sleep, as the sole recorder.

At the end of a reporting day, including possibly multiple hearings on different subjects, Reynolds typically erased it all this way: "When I go home at night, I sit on the porch, have a bourbon. I used to have a cigarette, too, but no more. I replay the day in fast-forward [his emphasis]. Sometimes I call the secretary's office and correct something in the record that I've recalled. Not often." (Reynolds, l992)


In summary, organizations and institutions know what they are doing--that is, they document their practices, articulate their reflexive understanding of them, and remember them--by keeping records. Once they exist, the records, in turn, enable others potentially to know about and to influence the organization's or institution's doings.

An example appeared recently in the news: a transcriber of cigarette industry meetings turned routine records of production processes over to a congressional committee. After seeing the records, the comittee held hearings to question industry executives about the manipulation of nicotine levels in cigarettes. The hearings were officially recorded by committee stenographers as well as broadcasted on C-SPAN. The executives' answers to the committee's questions during the hearing led Food and Drug Administration officials (sanctioned by the committee hearing record) to call for federal regulation of nicotine as a drug and cigarettes as drug-delivery devices. If governmental regulation of cigarette nicotine levels is adopted as social policy, the regulations will be written in reference to the published written transcript of these and related hearings, as the authoritative source of legislative intent. At all stages, this negotiated construction of social policy will have relied on record-keeping activity.

Thomas Jefferson was right. If governmental deliberation is to be remembered, revised, or acted on, it must first be recorded. Recording methods merit study. They are significant historical and material influences on processes and institutions of governance.

In response to scholarly claims that record-keeping functions as social control, I argue that record-keeping's potential for social construction is the prior issue, preceding effects of social control. As scholars, we can better assess the protean effects of institutional records if we consider their prior conditions or origin,including recording methods.

This commentary adapts C. Smith, l994a