In-Depth: A 500-Year History Of How To Set The Watch

In-Depth: A 500-Year History Of How To Set The Watch

“At the tone, the time will be … ”

Watches, it hardly needs to be said, need to be set to be of any use; and to be of any use, they must be set to a time standard that is trustworthy. This is something we scarcely think about today – someone is taking care of it; the Internet is taking care of it or atomic clocks are taking care of it or the US Naval Observatory is taking care of it. For most of the history of watchmaking, however, those resources weren’t available. and so other time standards had to be used if you wanted to set your watch precisely. One of my earliest watch-related memories is of dialing a telephone number to get the time – I did not, at age six or seven, yet own a watch but there was something fascinating about calling for the time and hearing a gentle voice say, “At the tone, the time will be … ” followed by an announcement of the time, and then, a short beep.

Night sky seen from the Paranal Observatory, Chile; celestial phenomena were the first time standard. Image, Wikipedia

In the early days of watchmaking – the end of the 16th century and much of the 17th – setting a watch was not a particularly emotionally stressful experience, because watches were not especially accurate anyway. Often they ran for less than a day, and if they kept time to within a half an hour a day, they were performing well. Serious timekeeping was a matter for public clocks, and for sundials. The latter provided probably the most accurate time, as the performance of a sundial is based on the rotation of the Earth, which is extremely stable. The Earth’s rotation is slowing ever so slightly thanks to tidal effects from the Moon, but it amounts to a minute lengthening of the day, which has only increased in length by 2.3 milliseconds per century (at least since the 8th century BCE, which is as far back as astronomical records allow us to make an analysis). At night, the time could be told with extreme accuracy by observing stellar transits, which are the times at which a given star crosses a certain point in the night sky – again, this is a measure of the stability of the Earth’s orbit, and thus an extremely reliable time standard. The ancient Egyptians, for instance, used a device rather poetically called a merkhet, which means, “instrument of knowing,” to observe the transits of stars, in order to tell the time at night. Needless to say, they were not using the merkhet to set watches (unless ancient Egyptian technology was far more sophisticated than we suspect) but the basic principle of using astronomy as most precise time standard continued for many thousands of years, up until almost the present day.

Merkhet, London Science Museum

Setting a watch, if you were lucky enough to own one in the 16th or 17th century – portable timekeepers were for the rich and often, even if a watch were cased in gold and precious stones, the movement would have been more valuable than the case – was not the obsessive matter that it is today. A sundial would suffice if the day were fine, or, after the development of the first pendulum clocks, a clock. The pendulum clock, as far as we know, was invented by the Dutch mathematician and physicist, Christiaan Huygens, in 1656 and patented the following year; his design was built by clockmaker Salomon Coster. After subsequent refinements to the design – including the invention of the deadbeat escapement by the Englishman Richard Towneley, in 1657, which was notably used for two precision regulators made by Tompion, for the Greenwich Observatory, in 1676 – the pendulum clock would be the most accurate timekeeping device, period, up until the beginning of the production of practical quartz clocks in the mid-20th century.

It is interesting for watch lovers to reflect on the fact that from a technical perspective, the most fundamental problems in precision timekeeping were solved by pendulum clocks by the end of the 17th century (there were continual refinements later, including during the 20th century, but all the basic elements were there long before) and that the watch would spend the next two centuries or so, basically trying to catch up.

Christiaan Huygens, in a portrait by Caspar Netscher; image, Wikipedia.

Public clocks in their earliest incarnations were often not much better timekeepers than watches, thanks to the erratic performance of the verge escapement (the earliest known type of clock escapement, other than some of the escapements used on sophisticated water clocks) but gradually, many of the earliest tower clocks had their verge escapements replaced with pendulums and anchor escapements. Later, high-precision, constant-force gravity escapements, such as the one used in the great clock in the Tower of Westminster, (better known as Big Ben) came into use. Other than astronomical observations, the tolling of the time by such clocks, and of course, their dials, provided the most reliable time standard for anyone wanting to set the time accurately on their watch.

One of the most interesting public time standards are so-called time balls, the first experience of which, for many of us, was probably seeing the ball drop at midnight over Times Square, New York, on New Year’s Eve. As it turns out, using the drop of a large ball as a signaling device is a – well, I suppose you could say, time honored way of transmitting the time visually; it has the advantage over a bell, of showing a precise moment in time. The system is quite old – supposedly the Greeks used the drop of a time ball as a time signal in the city of Gaza, after the death of Alexander the Great – and they were much used to transmit the time to ships at harbor from a station on land, so mariners could set their chronometers before setting sail. One of the best known time balls is on top of the Octagon Room at the Greenwich Observatory, and was installed in 1833; it has been dropping at 1:00 PM every day ever since (in the United States it is apparently customary to drop the ball at noon instead).

The Great Clock of Westminster (sometimes called Big Ben, though the name is really a nickname for the largest bell in the tower, not the clock itself). Image via Wikipedia.

Generally, the procedure seems to have been to start lifting the time ball five minutes or so before the hour, in order to alert the ships at anchor that the time signal was about to be sent; the time was recorded at the moment the ball began to drop, not when it reached the bottom. There are several dozen time balls still in operation around the world, including (to pick just one) the Deal Time Ball, in Deal, England (Deal was an important anchorage for fighting ships during the Napoleonic Wars). The Deal Time Ball was actually triggered remotely, by a telegraph signal from the Royal Observatory. In some locations, the time was signaled by firing a gun – these were set off at different times depending on the location, although noon has always been a perennial favorite. In Cape Town, South Africa, a noon gun is still fired as it has been every day at noon since 1806, from a battery on Signal Hill. There is also a famous 9:00 gun fired every night in Vancouver, British Columbia – this particular cannon is a muzzle-loader, a 12-pounder for you naval history buffs, which was cast in 1816 and has the crests of George III and the First Earl of Mulgrave, who was Master-General of Ordnance, on the barrel.

The Time Ball at the Royal Observatory.

The invention of the telegraph provided the ability for multiple locations to not only receive an accurate time signal more or less instantaneously, but also to be coordinated to the same time; this was especially essential for the railroads, the development of which spurred the evolution of time zones. The problem of setting a watch accurately had become much simpler, although a fundamental transition had occurred. For most of the history of timekeeping, the time in question was local time; to put it in simplest terms, noon was when the Sun stood at its zenith in the sky. Then, gradually, the concept of mean time evolved – mean time relies on the notion of a day of average length, which is the same throughout the year, rather than on the solar day, which can vary by a quarter of an hour or more either way, depending on the time of year. The difference between mean local time and solar time is in fact nothing other than the famous Equation of Time, and to aid in setting clocks by a sundial (in your country manor home, far from the madding crowd where the tolling of Big Ben would have been inaudible, perhaps) pendulum clocks often had tables for the Equation of Time pasted inside their cabinets, for easy reference.

The Vancouver 9:00 gun going off, in May 2017; image, Wikipedia.

The telegraph, time balls, and the dials and bells of telegraph controlled public clocks, began the era of widely available precision time, and thanks to the telegraph, railway station clocks became as precise and reliable a time standard as any other public clocks. Several inventions, however, combined to make accurate time references even more widely available: the telephone, radio, and that infamous corrupter of youth, the television.

The broadcasting of time signals over radio networks made telegraphy obsolete for both clocks and watches, radio stations began broadcasting time signals over general use radio networks, for the edification of the public, as early as 1922, when the BBC began the practice. In 1926 the Bulova Watch Company of New York broadcast the first radio advertisement with a time signal – ‘At the tone, it’s eight o’clock, Bulova watch time,” and on July 1, 1941, the very first TV ad was broadcast – this was also a Bulova ad, which was unlike much modern advertising, a model of concision and simplicity. Through the miracle of the Internet we can still enjoy this advertisement on YouTube.

Accurate time standards by which to set one’s watch continued to become more and more ubiquitous. My earliest memory of such a standard, designed to enable anyone to accurately set their watch or clock, was simply calling a time service. In 2016, The Atlantic’s Adrienne LaFrance wrote a wonderful reminiscence about using a telephone to call for the time; she writes, in part:

“The first two telephone numbers a little kid learns are usually 911 and their own.”

“When I was about 5 years old, I decided to call 911—just to see what would happen, even though I knew I wasn’t supposed to. When a dispatcher answered, I promptly hung up. They promptly called back.”

“This led to a conversation I suspect many mothers have had with many curious children. It must have been around the same era when I learned of what seemed at the time to be a great secret of telephony. There was another number I could call, and I could call it any time and as many times as I wanted, without getting in trouble.”

Bell Telephone System international switchboard, 1943.

The number varied depending on your location, but the result was generally a soothing voice, saying pleasantly, “At the tone, the time will be,” followed by the time, and then an equally pleasant beep. Various automated systems evolved over the years and surprisingly, LaFrance reports, quite a few people still like to call and get the time; interviewed for the story, the chief scientist for time services at the US Naval Observatory, Demetrios Matsakis, said that they get around 3 million calls per year. If you ring up the USNO time service number you will hear the time given by a voice that sounds straight out of Mission Control circa 1965. That voice is none other than Mr. Fred Covington (1928-1993) known for such films as The Bingo Long Traveling All-Stars & Motor Kings, and The Battle Of Chickamagua (narrator). He became the voice of USNO time in 1978, and though he has passed since then into that realm wherein Time giveth way to Eternity, you can still hear his voice, reciting the time on demand, by dialing 202-762-1401.

The HODINKEE app’s Time screen: UTC, perpetual calendar, moonphase, and of course, the time.

Electronic time signals – first from the telegraph, then the telephone, radio, and TV – made things like the time ball, cannon, and public clocks obsolete as practical time references; they exist today for cultural and historical reasons, not because we need them to set our watches. The necessity for a daily accurate time reference, already considerably diminished by ongoing improvements in mechanical watch technology during the 20th century, decreased further with the advent first of tuning fork watch technology (in the Accutron and other tuning fork movements) and with the appearance of quartz watches, in 1969, which made timepieces accurate to within a minute or less per month, more or less ubiquitous. Still, owners of wristwatches, quartz or mechanical, still need to set their timepieces occasionally, and lots of us enjoy checking the accuracy of our mechanical watches against a reliable reference.

For many the time displayed on their cell phone is more than accurate enough, and for good reasons. In order for the Internet to function, it’s necessary for system clocks in mobile devices and computers to be synchronized precisely, across the entire network. The Network Time Protocol is designed to synchronize participating devices to within a few milliseconds of UTC, which is a time standard that is, in turn, regulated by International Atomic Time, which is an average of the time kept by over 400 atomic clocks in over 50 national labs around the world. (Many modern electronic watches can obtain the time from GPS or radio signals, which are also regulated by atomic clocks and such watches effectively never need to be set, except perhaps for local transitions to and from Daylight Saving Time). The HODINKEE app uses the time generated by the system clock on the iPhone. Originally both the HODINKEE and Watchville apps synchronized directly with Internet time servers via NTP, but with the release of iOS 9, our developers noticed that the amount of drift between NTP and the iPhone’s clock had decreased significantly – so much so that the apps no longer need to obtain the time directly from NTP.

No external standard needed? The Citizen Caliber 0100, accurate to one second per year.

The act of setting a watch is an interesting one philosophically and psychologically. You are attempting to make your watch as accurate as possible but in setting it to a time signal or other time standard, you are also acknowledging the limitations of the watch itself. Watches keep getting more and more accurate, however, which is a pleasant surprise if you are afraid that the art of watchmaking is stagnating. Several companies – Omega, Rolex, Grand Seiko, Citizen, and others – continue to invest considerable time, energy and effort into making more accurate watches; in fact Citizen, earlier this year, announced the most accurate and most precise watch ever made, in over 500 years of watchmaking, so clearly time and progress both march on, at least as far as watches are concerned.

As watchmaking improves, access to an atomic-clock accurate standard becomes less of a necessity, and more a way of enjoying and observing the degree to which one’s watch, or watches, precisely measure the passage of time – and how close we can get to the centuries-old dream, of set it and forget it.

Share Post
Written by
Latest comment
  • Wow What a great article. Very interesting and informative! I to can remember calling the time as a kid, I had totally forgotten about that! I also just called the current number in Washington DC, and yes, heard Mr. Covington tell me the exact time that My Citizen pro master also revealed. Thank you for the information and the trip down memory lane!

LEAVE A COMMENT

css.php