Inventing history:
Garrett Morgan and the traffic signal
In early 1922, African-American inventor Garrett Augustus Morgan designed a cross-shaped traffic signal, for which he submitted a patent application on February 27 of that year. The patent — which was not even among the first 50 traffic signal patents issued in the United States — was granted on November 20, 1923. For whatever reason, numerous writers and public figures have credited Morgan with inventing any or all of the following:
- world's first traffic signal
- first traffic signal to earn a patent
- first automatic traffic signal
- first traffic signal with a third "all-directional stop" phase
- first signal with a yellow light phase
- the basis for modern traffic signal systems
None of these claims are even remotely true, as rest of this page shows.
Myth: Garrett Morgan invented the first traffic signal
Some notable early signals, prior to Morgan's 1922 invention
London, 1868
The first known signal device for regulating street traffic was installed in 1868 in London, at the intersection of George and Bridge Streets near the Houses of Parliament. Designed by railroad signal engineer JP Knight, it had two semaphore arms which, when extended horizontally, meant "stop"; and when drooped at a 45-degree angle, meant "caution." At night, red and green gas lights accompanied the "stop" and "caution" positions (Sessions 1971; Mueller 1970).
By the signal "caution", all persons in charge of vehicles and horses are warned to pass over the crossing with care and due regard to the safety of foot passengers. The signal "stop" will only be displayed when it is necessary that vehicles and horses shall be actually stopped on each side of the crossing, to allow the passage of persons on foot; notice being thus given to all persons in charge of vehicles and horses to stop clear of the crossing.
Proclamation of Richard Mayne, London Police Commissioner, in 1868; quoted in Mueller 1970
Salt Lake City, about 1912
A contender for "inventor of the first electric traffic light" is Lester Wire of Salt Lake City (Sessions 1971).
Mr. Wire, who died in 1958, was a Salt Lake police officer who invented the first electric traffic light in 1912.... The first hand-made model was a wooden box with a slanted roof so rain and snow would fall off. The lights were colored with red and green dye and shone through circular openings. The box was mounted on a pole, and the wires were attached to the overhead trolley and light wires. It was operated by a policeman. In ensuing years, Mr. Wire improved upon the first model.
"Peak Named for Inventor," Deseret News (Salt Lake City), February 9, 1967
Cleveland, Ohio, 1914
On August 5, 1914, several years before Garrett Morgan invented his T-shaped semaphore-type signal, the American Traffic Signal Company installed red and green traffic lights at each corner of the intersection of 105th Street and Euclid Avenue in Cleveland (see this 1914 Motorist article documenting the debut of the lights).
The installation was patterned after the design of Cleveland inventor James Hoge (Sessions 1971; Mueller 1970), whose U.S. patent #1,251,666 describes a system of electrically powered stop-go indicators, each mounted on a corner post. In Hoge's design, the signals are wired to a manually operated switch housed inside a control booth, and are electrically interlocked in such a way as to make conflicting signals impossible. Also described in the patent is a system to allow communication between the signal controller and the police and fire departments. The Cleveland installation incorporated all of the above elements in some form or other, plus a bell to warn the drivers of color changes.


from Hoge's patent #1,251,666
William Potts' 4-way, red-yellow-green signals, Detroit, 1920
In October and December of 1920, a Detroit policeman named William Potts constructed several red-yellow-green light signal systems. Some lights were mounted atop "traffic towers" manned by policemen; others were overhead suspension lamps remarkably similar in form to a modern traffic light. The 4-direction traffic lamp pictured below is of the latter type, and is on display at the Henry Ford Museum in Dearborn, Michigan.

A 1920 vintage Potts signal, at the Henry Ford Museum.
William Potts
An entry in the museum's artifact database describes the item:
The world's first three-color, four-direction, elec. traffic lamp, was installed at the intersection of Woodward Ave. and Fort Street, Detroit, Michigan in October, 1920. It was designed by Superintendent (then inspector) William L. Potts of the Signal Bureau, Detroit Police Department. Basic design remains practically unchanged today. The signal remained in use until 1924 and became a part of the world's first synchronized signal system. This system extended from Jefferson to Adams on Woodward Avenue and was controlled manually from a tower at Woodward and Michigan.
museum archives, Henry Ford Museum & Greenfield Village (as of Jan 29 2003)
The meanings of the colored lights were essentially the same as today. Green meant "go"; red meant "stop"; and yellow (amber) meant "clear the intersection" (Mueller 1970). An analogous color scheme had been used by the railroads, where as early as 1899, the New York, New Haven and Hartford Railroad introduced a system wherein red, yellow, and green meant "stop", "caution", and "all clear" respectively (Brignano 1981).
For more about Potts and his signals, read this 1947 Motor News article entitled "Mr. Trafficlight".
New York City, early 1920s
In New York City, Dr. John F. Harriss, Special Deputy Police Commissioner in charge of traffic control, organized a system of red, yellow, and green lights to control traffic along Fifth Avenue. The idea reached the experimental stages as early as February 1920 (New York Times, Feb 06 1920). By March of that year, the colored lights were in service ("The Traffic Lights," New York Times, Mar 15 1920).
By January 1922, an interconnected matrix of traffic lights was expanding throughout Manhattan:
Dr. John F. Harriss, Special Deputy Police Commissioner, began experimenting yesterday with powerful signal lights which will be installed from week to week until traffic in most of Manhattan will be simultaneously stopped and started by red, green and yellow lights all operated by a single switch in Times Square.
"To Rule All Traffic from Times Square", New York Times, Jan 05 1922, p.1
The original New York traffic towers were not true four-direction signals like the lights in Detroit. Instead of simultaneously shining different colors in perpendicular directions, the Manhattan signals shone only one color at a time: red for north-south movement (main avenues), yellow for all traffic to stop, and green for east-west movement (side streets). Stationed at each tower was a traffic officer to enforce the signals.
Myth: Morgan's signal was the first to earn a patent
In the U.S. alone, more than 60 traffic signals patented before Morgan's
A common variation of the Morgan myth stops short of crediting Morgan with the first-ever traffic signal, instead bestowing upon him the honor of supposedly having received the first patent for one in 1923. Actually, the earliest US traffic signal patent was issued to Ernest E. Sirrine in 1910, and dozens more followed in the years preceding Garrett Morgan's invention.
| Patent# | Inventor | Invention Title | Filed | Patented |
|---|---|---|---|---|
| 976,939 | E.E. Sirrine | Street Traffic System * | 4/28/1910 | 11/29/1910 |
| 991,964 | T.M. Flaherty | Signal for Crossings | 9/24/1909 | 5/9/1911 |
| 1,084,411 | S.F. Du Ree | Traffic Lantern | 4/1/1913 | 1/13/1914 |
| 1,123,925 | R.W. Roderick | Traffic-Guide | 1/27/1914 | 1/5/1915 |
| 1,158,610 | O.H. Whitman | Crossing Signal | 7/20/1914 | 11/2/1915 |
| 1,181,719 | A.A. Anderson | Traffic Control Semaphore | 9/1/1915 | 5/2/1916 |
| 1,184,598 | J.G. Wallmann | Apparatus for Regulating Street Traffic | 5/20/1914 | 5/23/1916 |
| 1,195,583 | F.B. Henretta | Traffic-Director * | 5/26/1916 | 8/22/1916 |
| 1,224,632 | W. Ghiglieri | Traffic Signal * | 1/12/1915 | 5/1/1917 |
| 1,232,375 | J.W. Nethercutt | Street-Crossing Signal | 1/13/1917 | 7/3/1917 |
| 1,234,249 | W. Zabel | Signal | 9/5/1916 | 7/25/1917 |
| 1,236,441 | F.J. Husbands | Automatic Traffic Signal * | 10/9/1916 | 8/14/1917 |
| 1,236,625 | J. Tomko C. Oppenlander | Electric Signal Device | 12/13/1915 | 8/14/1917 |
| 1,240,090 | E.A. Pound | Automatic Traffic Regulator * | 7/11/1916 | 9/11/1917 |
| 1,240,640 | H.T. Winston | Traffic Signal | 12/21/1916 | 9/18/1917 |
| 1,249,539 | F.D. Spear | Traffic Signal | 10/6/1916 | 12/11/1917 |
| 1,250,083 | J.A. Black | Traffic Signal | 8/3/1917 | 12/11/1917 |
| 1,251,666 | J.B. Hoge | Municipal Traffic Control System | 9/22/1913 | 1/1/1918 |
| 1,254,305 | C.W. Blakeslee | Street Traffic Signal | 1/24/1916 | 1/22/1918 |
| 1,254,862 | H.C. Strout | Traffic Sign | 3/3/1917 | 1/29/1918 |
| 1,254,971 | W.L. Bradley | Traffic Signaling Device | 11/8/1915 | 1/29/1918 |
| 1,259,350 | W.A. Braden | Crossing-Signal | 7/25/1910 | 3/12/1918 |
| 1,261,251 | C.A. Lynch | Electrically-Operated Crossing-Signal | 9/19/1916 | 4/2/1918 |
| 1,264,114 | F. Milliken | Street Light [signal] | 8/18/1917 | 4/23/1918 |
| 1,264,300 | H.D. Heuer | Street Traffic Signal Apparatus | 5/23/1914 | 4/30/1918 |
| 1,267,767 | B.M. Harris | Street Traffic Controlling Apparatus | 1/25/1915 | 5/28/1918 |
| 1,268,288 | J.W. See | Signal for Street Intersections * | 7/8/1914 | 6/4/1918 |
| 1,273,607 | G.R. Hawks | Traffic Signal | 9/20/1916 | 7/23/1918 |
| 1,275,749 | S.C. Repsholdt | Traffic Signal Tower | 9/2/1916 | 8/13/1918 |
| 1,283,858 | F. Milliken | Street Lamp [signal] | 11/17/1917 | 11/5/1918 |
| 1,295,585 | C.H. Peck | Traffic Guide | 8/3/1915 | 2/25/1919 |
| 1,301,396 | W.A. Dau | Traffic Signal | 4/19/1916 | 4/22/1919 |
| 1,307,544 | O.A. Erdmann | Traffic-Signal * | 11/29/1915 | 6/24/1919 |
| 1,308,658 | W.E. Carr | Street-Crossing Signal | 9/20/1916 | 7/1/1919 |
| 1,313,445 | F.A Wright, W.J. Brong | Electrical Grade Crossing Signal | 6/1/1917 | 8/19/1919 |
| 1,315,049 | D.C. Stickrath | Portable Street Signaling Device | 3/29/1919 | 9/2/1919 |
| 1,317,009 | L.P. Ernst | Traffic Signal Control | 1/28/1919 | 11/23/1919 |
| 1,317,819 | C.H. Peck | Portable Street Traffic Signal | 3/8/1916 | 10/7/1919 |
| 1,326,009 | R.E. Wetter | Signal Apparatus | 1/20/1919 | 12/23/1919 |
| 1,328,269 | B.W. Davis | Signal System for Controlling Street Traffic | 10/30/1914 | 1/20/1920 |
| 1,331,574 | L.F. Maire | Traffic-Signal | 4/12/1917 | 2/24/1920 |
| 1,331,691 | G.W. Bennett | Traffic Sign * | 12/23/1918 | 2/24/1920 |
| 1,332,917 | C.H. Peck | Street Traffic Signaling Device | 4/19/1915 | 3/9/1920 |
| 1,335,240 | F.J. Husbands | Traffic-Signal * | 3/11/1918 | 3/30/1920 |
| 1,338,028 | D.J. Linehan | Traffic-Signal | 5/15/1919 | 4/27/1920 |
| 1,353,999 | T.O. Miller | Street Traffic Signal | 11/26/1917 | 9/28/1920 |
| 1,354,249 | D.W. Gore | Traffic Signal | 12/6/1919 | 9/28/1920 |
| 1,355,361 | J. Selligman | Traffic Signaling Device | 6/17/1920 | 10/12/1920 |
| 1,356,276 | J.H. Gourley | Highway-signal | 4/8/1919 | 10/19/1920 |
| 1,357,875 | A.W. Kent | Traffic Signaling System | 3/5/1917 | 11/2/1920 |
| 1,359,529 | S.C. Repsholdt | Traffic Signal Tower | 7/15/1919 | 11/23/1920 |
| 1,360,796 | S.C. Repsholdt | Traffic Signal Tower | 7/15/1919 | 11/30/1920 |
| 1,368,185 | J.E. Miller | Traffic Director | 5/28/1920 | 2/8/1921 |
| 1,372,401 | E. Cadeddu | Street Signaling Device | 9/11/1920 | 3/22/1921 |
| 1,382,482 | J.W. Curry | Traffic Directing Apparatus | 10/21/1918 | 6/21/1921 |
| 1,385,807 | R.E. Wetter | Traffic Signal System | 11/23/1918 | 7/26/1921 |
| 1,394,918 | L.C. McAdams | Traffic-Signaling Mechanism | 1/26/1920 | 10/25/1921 |
| 1,394,966 | J.W. Brenkert | Street Traffic Signaling Device | 3/27/1920 | 10/25/1921 |
| D59,785 | G.N. Russel, O.W. Russel | Traffic Signal (design) | 7/5/1921 | 11/22/1921 |
| 1,430,901 | J.A. Hamberg | Traffic Signal | 7/12/1921 | 10/3/1922 |
| 1,434,050 | A. Duran | Semaphore | 7/13/1920 | 10/31/1922 |
| 1,447,659 | T.E. Hayes | Combination traffic guide and traffic regulating signal | 8/7/1922 | 3/6/1923 |
| 1,448,379 | H. Bock | Traffic Signal * | 11/8/1921 | 3/13/1923 |
| 1,462,266 | G.C. Degner | Traffic Regulator * | 11/3/1920 | 7/17/1923 |
| 1,465,346 | H.W. Chittenden | Traffic Signal | 8/26/1922 | 8/21/1923 |
| 1,466,809 | Lola Strong | Automatic Traffic Signal * | 11/16/1920 | 9/4/1923 |
| 1,467,206 | W.F. Stellner | Traffic Signal | 10/12/1922 | 9/4/1923 |
| 1,467,303 | H. Bezer | Highway-crossing protection | 8/28/1918 | 9/11/1923 |
| 1,473,448 | C.M. Snell | Automatic Traffic Signal * | 5/23/1923 | 11/6/1923 |
| 1,475,024 | G.A. Morgan | Traffic Signal | 2/27/1922 | 11/20/1923 |
Asterisk (*) denotes an automatic signal (which, for the purposes of this table, means having a standard or optional capability of operating by itself, without requiring human control or vehicle actuation).
Not all new signals were patented, especially those created as public projects by police departments (such as in Salt Lake City, Detroit, and New York City).
Myth: Morgan's signal was automatic
It was operated with a hand crank
Certain website authors, while conceding that other traffic signals preceded Morgan's, try to sell yet a different variation of the story: that Morgan patented the very first automatic traffic signal. Besides ignoring several automatic signals preceding the Morgan patent (see patent table above), the originators of this myth apparently did not read the patent itself, which describes a mechanical, crank-actuated unit operated by a human traffic director who must turn the crank every time the signal is to change from "stop" to "go" and vice versa:
The operation of a signal constructed according to my invention is as follows:—Assuming that traffic is moving in the direction indicated by the “Go” characters in Fig. 1 and that the director desires to change the right of way, then the crank 28 is turned, whereupon the arms 13 are immediately raised by virtue of the rack and pinion connections actuated by the crank shaft 27. ... Then as the crank is turned still further, the gear 32 engages the gear 33, whereupon the indicators are revolved while the arms are still held in vertical position. ... [B]y the time the crank has been turned a complete revolution, the arms are in substantially horizontal position and are automatically held in the desired direction by the indexing plunger 36.
from Garrett Morgan's US Patent No. 1,475,024 (emphasis added)
Myth: Morgan invented the first three-color or three-phase signal
Garrett Morgan's 1922 invention was by no means the first traffic signal to use a third signal phase in addition to the usual stop and go. The Detroit and New York police departments had been using three-color traffic lights since 1920. The rest of the country would eventually follow suit, adopting red, yellow, and green lights to control road traffic.
The Morgan signal had no yellow light. Instead, it had a "third position" that displayed the word STOP in all directions before allowing traffic to proceed in any one direction, thereby providing extra time for the intersection to clear. This method of including an "all-stop" phase as part of the regular signal cycle goes back at least seven years earlier to January 1915, when William Ghiglieri of San Francisco described the operating cycle of his own red-green traffic light:
First, a short series of rings from the bell accompanied by a green display east and west and a red display north and south, this starting the east and west traffic. Second, a change to red on all four sides without a bell signal, thus stopping the east and west traffic. Third, two short series of rings from the bell with green displays north and south and red displays east and west, this starting the north and south traffic. Fourth, a change to red on all four sides without a bell signal, thus stopping the north and south traffic. This cycle is then repeated.
William Ghiglieri's US patent #1,224,632 (emphasis added)
Application filed January 1915; patented May 1917.

Ghiglieri signal
(cutaway view)
It is noteworthy that Ghiglieri's signal — invented seven years prior to Morgan's — could do all that the Morgan signal could do; furthermore, it could be controlled by automatic timer whereas Morgan's could not; and on top of that, Ghiglieri's alternating red-green light design bore far more resemblance to a modern traffic light than did Morgan's mechanical semaphore.
The all-red interval was rarely if ever used in three-color signals until 1950, when the Institute of Traffic Engineers began recommending that a momentary all-red phase follow the yellow phase (Eccles 2001).
Myth: The Morgan signal was important or influential
Despite claims on various websites that Morgan's invention was used "throughout North America," the absence of his signal in 1920s photographs and news articles suggests that it was not installed in large numbers anywhere. Notably, it did not merit a single mention in the book-length historical study by Gordon M. Sessions (1971), which covers a wide variety of devices in tracing the development of traffic control.
The oft-repeated story that Morgan sold his patent to General Electric for $40,000 — an enormous sum back then — should be taken with a grain of salt. No record of this transaction appears either in the US patent assignment records at the National Archives, or in the GE historical business records at the Schenectady (New York) Museum, or in Morgan's legal and business papers at the Western Reserve Historical Society in Cleveland. Ads and photos from the 1920s indicate that GE's early traffic signal products were of the type with a row of colored lights, not the semaphore type.
William Potts' 1920 invention
Garrett Morgan's 1922 invention
Although many prefer to think that the signal on the right formed the basis for modern traffic control systems, the less inspiring reality is that the traffic light had already evolved into its familiar red-yellow-green form before Garrett Morgan came along with his T-shaped semaphore. Technologically, the Morgan device was more dead end than prototype: its crank-driven design did not lend itself to the automation and interconnection that was already happening with the fully electrical signals of the same era. It was novel enough to earn a patent, and was admirable and creative, and so forth, but contributed nothing new in terms of functionality and played no part in the chain of development that led to the modern traffic light.
References
US Patent Database (http://www.uspto.gov/patft/index.html)
Gordon Sessions, Traffic Devices: Historical Aspects Thereof, (Washington DC: Institute of Traffic Engineers, c.1971).
Edward A. Mueller, "Aspects of the History of Traffic Signals", IEEE Transactions on Vehicular Technology, vol. VT-19, no. 1, pp.6-17 (1970).
M. Brignano and H. McCullough, The Search for Safety: A History of Railroad Signals and the People Who Made Them (American Standard, Inc. 1981)
Eccles and McGee, "A History of the Yellow and All-Red Intervals for Traffic Signals" (Washington DC: ITE, 2001)
Gas mask
In regard to equally erroneous claims that Garrett Morgan invented the first smoke mask or gas mask in 1914, or that it became the gas mask used by U.S. soldiers in World War I, see The Invention of the Gas Mask and the references therein.