‘A plea for uniformity’: time services in London

In 1909 H. G. Wells described London as ‘the Imperial city – the centre of civilisation, the heart of the world’,Footnote ¹³ reflecting an enduring notion of the centre controlling the peripheries. The status of London as the capital city caused much discussion over a protracted period about whether and how its time should be standardized and synchronized. ‘A plea for uniformity’ was the cry from the chairman of STC in 1908 in a lecture to City of London elders and business-folk, part of a long-running public debate on ‘lying clocks’ that appeared to resonate with a social and economic culture of ‘synchrony’. Sir John Cockburn of the British Science Guild wrote to The Times asking for ‘censorship’ of the time kept by public clocks in London: ‘Highly desirable as individualism is in many respects, it is out of place in horology. A lying timekeeper is an abomination, and should not be tolerated. A bye-law might well be framed requiring clocks in public places to be synchronized with standard time.’Footnote ¹⁴ Access to precise standard time was seen as the sine qua non of a modern metropolis and London's status in the urban hierarchy was held to rely on its ability to capitalize on developments in standardizing time. Germany, ‘our great commercial rival’, and Berlin in particular, which had large time-distribution systems, were held out as examples of what London and by extension the empire was missing.Footnote ¹⁵ Yet this 1908 debate was hardly new. Such scientific bodies as the Royal Observatory (RO) and the Royal Astronomical Society (RAS) had long highlighted the concern. Robert Lecky, FRAS, having spoken to the Astronomer Royal, wrote in 1879 to the president of the RAS to bring before him

The establishment in 1876 of STC to provide standard time signals hourly across a telegraph-wire network in London contributed an important voice in this standardizing rhetoric. As a business it operated as late as the 1960s. Its main competitor (ironically, its supplier of Greenwich time) was the public telegraph service provided by the General Post Office (GPO) in direct connection with the electric clocks at the RO, which was an extension of the RO's network of time-signalling for the Admiralty, railway companies and the GPO's own use. The users of STC's telegraph time service were many and varied, according to their subscriber lists in the 1880s.Footnote ¹⁷ Many were businesses whose relationship with time was fairly clear. The standard account cites railways as the primary driver for the adoption of standard time, owing to their requirements for a common time in timetabling and signalling arrangements. Yet the story does not end there, important though railways undoubtedly were. Banks and other financial institutions carried out monetary transactions which required time-stamps and therefore a coherent and agreed form of standard time. STC's nascent time-distribution system was reported in The Times in 1877 with the news that ‘The Bankers' Clearing-house immediately availed themselves of the advantage of a system which, being placed in communication with the Lombard-street and other leading banking establishments, will remove the often-occurring disputes as to “paying in” within the specified time.'Footnote ¹⁸ Watch- and chronometer-makers needed time to adjust and set their products before despatch to customers. For chronometer-makers, whose clients were usually naval officers in the Royal Navy or merchant service with their entire ship and crew dependent on the good running of the chronometer, since accurate longitude determination depended upon the instrument, the demand for accurate time was one of potentially life-or-death magnitude. The British Horological Institute (BHI), the trade body representing the clock and watch trade, wrote to the Astronomer Royal in 1879 asking for his help in setting up a time-distribution system: ‘The Council of the Institute are exceedingly anxious to fit a clock always showing correct time, or rather shewing the same time as the standard clock at the Greenwich Observatory. To be able to refer to such a clock would be of great value to chronometer-makers and others.’Footnote ¹⁹ George Airy, the Astronomer Royal, agreed, saying that ‘the maintenance of a clock shewing true time, at the B.H. Institute, is very desirable, and I shall be pleased to assist in it’.Footnote ²⁰ One can note here the subtle difference between ‘true’ or ‘correct’ time, and the time at the Greenwich Observatory, and that this distinction was recognized by the BHI but ignored by the RO. The tension between time-of-day and time-everyone-can-agree-upon underscored much of the debate over regulation and standardization in this period.

To agree upon and encourage trust in a standard, it was crucial to understand how that standard was disseminated. What route had the standardized measure taken from its definition to its (many) realizations? An example: in 1894 a correspondent wrote to the Astronomer Royal, William Christie, saying, ‘I should be much obliged to you if you would tell me if the hour signal at Websters No. 5 Queen Victoria Street EC which professes to be given by electric current from Greenwich is to be depended upon.’Footnote ²¹ The Royal Observatory replied, ‘Webster gets his hourly time signal from the Normal clock of the Standard Time Co, who correct their clock by the 10 am signal from Greenwich. Webster's signals are therefore affected by any error of the Standard Time Co.'s clock. Hourly signals direct from Greenwich … are supplied by the General Post Office.'Footnote ²² Standardization relies on hierarchies of authority and trust. Any standard measure comprises a definition, a primary ‘master-standard’ reference and lower-order secondary-standard references. To speak of ‘standard time’, therefore, was to address a complex chain of artefacts, systems, owners and operators, commencing, for the purposes of this discussion, in the mean solar standard clock at the Royal Observatory, Greenwich.Footnote ²³ Whilst members of the public may not have fully grasped the complexity of this pyramidal system, STC certainly had, and sought to extract public-relations value by borrowing the authority of the Royal Observatory. They presented their time signal as being synonymous with Greenwich time. Unfortunately for them, though, the Royal Observatory, at the top of the pyramid, was just as aware of the publicity value and sought instead to consolidate its own position by distancing itself from the lower-order standards. It began in earnest in 1901, when STC wrote to William Christie at the RO, bringing to his attention the firm's time service, then of twenty-five years' standing:

He proceeded to request a monthly certificate of STC's clocks' error, but Christie's own initial response, pencilled onto the letter upon receipt, was cautious: ‘Mr Lewis, I don't think we should do this for a private concern.’ He then evidently had his assistant Lewis brief him about STC's service. Lewis's response – backing Christie's own assessment – demonstrates a concern not with accuracy but with opportunity. STC, he said, which had presented one of its corrected clocks to the RO a few years previously,Footnote ²⁵

The official response to STC was a refusal, but, naturally enough, did not mention money, stating,

Newitt, the managing director of STC, took some time before responding to this curt rejection, but when he did, in a long letter expanding on his company's work, he addressed both the issue of public interest explicit in the RO's rejection, and the issue of profit that lay behind the decision. In a rehearsal of the 1908 ‘lying clocks’ debate around public timekeeping, he argued that in the absence of government or municipal interest in time supply, Londoners would have no access to GMT without STC's service: ‘the Westminster Clock cannot be read nearer than half a minute’. The firm's service corrected a publicly visible dial at London's Guildhall, as well as being taken by clockmakers, ‘who, besides using it for comparing their own regulators, generally exhibit it in their windows as Greenwich Observatory Time for the public benefit’. Moreover, STC's was an extensive network: ‘Our time signals and synchronized Clocks are distributed between Ealing and Whitechapel, Islington and Kennington, and we claim that directly or indirectly nearly every Clock in London is regulated from our time signals.’

This was an audacious claim from STC's Newitt, who felt sufficiently emboldened to attack what he (quite correctly) felt was the true reason behind the RO's rejection: ‘The fact that … we may derive some profit, we submit should not deter you from benefiting the Public through us, particularly as the expense would be in addition to our expenditure, the effect would be negative.’ Of course the cost of the proposed certification service itself was not the problem, rather the added value gained by STC with state certification, so Newitt ended with what he hoped would be a clinching argument. He asked for a personal interview with the Astronomer Royal, ‘having in view the fact that the labour and expenses involved in complying with our request would be much less than that of rating chronometers, which we believe is undertaken by the Observatory, for makers who derive profit from their sale’.Footnote ²⁸ This was a mistake that lost Newitt the upper hand. The RO responded that the Astronomer Royal

There the correspondence ended.

It might be amusing to speculate why STC wanted certification from the RO in the first place, given correspondence earlier in 1901 that provides further evidence for an argument advanced elsewhereFootnote ³⁰ that the RO/GPO service was far from inherently accurate or reliable. On 4 March 1901 the GPO's engineer-in-chief wrote to the Astronomer Royal to discuss a letter he had received from STC reporting errors in the GPO time signal of up to several seconds, lasting for several days.Footnote ³¹ The RO replied with a catalogue of misfortune surrounding its master time-signal clock and the people operating it: a signal which ‘through an accident went out 9 seconds wrong’; a signal sent at 9 a.m. on 27 February ‘when the clock had dropped 4 seconds since the 1pm signal on the previous day & again at 9am on Monday March 4, when the clock had dropped 37 seconds since 1pm on Saturday, March 2’. On ending this tale of woe, the RO admitted, ‘the clock is now being cleaned’.Footnote ³² These episodes, notwithstanding their comedy of errors, offer evidence in line with Graeme Gooday's thesis of a division in this period between an older culture of measurement ‘labour’ (represented here by the work at the RO) and a newer one of ‘mechanization’ (attempted by STC). Whereas the RO's technology was openly human-resource-intensive (the correspondence in these files describes the practices of staff operating the time-signal equipment), the STC sought to make invisible the human agency, elsewhere describing in detail the automated, mechanized, fail-safe technological devices in use on their network that apparently required little human intervention. The role of trust in this context, in people and systems, is important.Footnote ³³

Overtime and calling time: Victorian reform and time

The records of STC provide valuable evidence for a wider context to this story of standardized time beyond the railways and the watchmakers and the rhetoric of the time-distribution marketing men. Two examples emerge clearly where standard time – and the Standard Time Company, to an extent – sat squarely within a culture of Victorian reformist legislation, concerning two of the most significant social concerns of the period: industrial working conditions and the perils of alcohol.

In 1902 the British socialist reformer, Sidney Webb, opined,

Besides reminding us of the relative modernity of these other now-entrenched institutions of daily life, Webb establishes the high status of workplace reform in the nineteenth-century canon of social regulation and standardization.Footnote ³⁵ Through much of the reform, time was the tool by which work would be regulated.

The Factory Act of 1802 is held to be the first of many reforming legislative acts regulating the hours and conditions of labour in certain factories for certain sectors of the working population. Subsequent acts amended working-hours legislation. The 1831 act, for instance, required that a register of actual working hours be kept by the employer and shown to the justices, at that time the enforcers of the acts, on demand. This act only applied to cotton mills. The 1833 act replaced justices as law enforcers with newly appointed factory inspectors. They could make enquiries, summon witnesses and levy fines following unannounced factory inspections. As well as inspecting factories, factory inspectors were obliged also to report their activities and findings to government. These reports provided politicians with crucial evidence of the actual state of affairs in factories and the true effects of the factory legislation. The 1844 act was a direct result of the inspectors' reports. It was in this act that a first time standard was defined for the regulation of working hours. It stated, ‘The hours of the work of children and young persons … shall be regulated by a public clock, or by some other clock open to public view’, said clock to be approved by the district factory inspector.Footnote ³⁶ The effect of this clause was summarized by later commentators thus: ‘These regulations amounted to a recognition, if a tardy one, of the need for defining as well as limiting the working day, and they made the inspectors’ work much more practicable.'Footnote ³⁷

Whilst 1844 saw legislation passed that defined a timescale by which all employers were bound, the idea was not new. In 1832 Parliament had debated Michael Thomas Sadler's factories legislation bill. In the bill Sadler described a known fraudulent activity:

Such a speed clock can be seen as a later example of a machine that measures variable, value-based units within what Ken Alder has termed a ‘local metrical dialect’, highlighting the complexity and contingency of measurement cultures. Here, that flexibility was deemed a problem. In the subsequent parliamentary debate Sadler explained that no matter how specific the legislation was in defining the hours that employees (in this case, children) were permitted to work, being ‘on time’ meant nothing if the time standard itself were open to ambiguity: ‘the child is not always safe, however punctual; for in some mills two descriptions of clocks are kept, and it is easy to guess, therefore, how they are occasionally managed’.Footnote ³⁹ The bill proposed defining a particular clock as the timekeeper for working hours, but before it could be passed a general election lost Sadler his seat. A further bill, introduced in 1838 by Home Secretary Fox Maule, included a similar clause but was not passed into law.Footnote ⁴⁰

Over half a century after the 1844 act things had hardly changed. Compare it with the Factory and Workshop Act, 1901:

And this was the extent of it. No mention of Greenwich time, no mention of a common time standard across the UK. Even contemporary analysts were content with the definition as it stands here. In a synoptic overview of factory legislation written in 1926, H. A. Mess writes about objective standards replacing subjective ones, and it is important to note here how he equates an instrument with a standard:

It is certainly true that clocks capable of accurate timekeeping were readily available during the nineteenth century. But time standardization is concerned with the selection of a standard by which to set other clocks. A clock might keep time precisely (it shows an hour has passed when the Earth has rotated fifteen degrees) and yet be wholly inaccurate (in other words, telling the wrong time of day – it might read 6 a.m. when the sun is directly overhead for noon). The factory clocks were only truly accurate if also set to a time standard, and it was the absence of an agreed standard that meant employers could tamper with the clocks or operate two clocks showing different times.

This is not just a philosophical point. Access to a common time standard for setting the designated factory clocks to comply with factory legislation was a practical desideratum. An example comes as late as 1913 in a request from the Oldham Master Cotton Spinners' Association to the GPO for a telegraphic time service. It is interesting that it was from a body in that particular industry, for if we return to Sidney Webb's analysis of factory legislation written in 1901, we see that he, like Mess, saw only problems already solved:

The cotton-spinners may well have enjoyed an enviable standard of life, but the ‘common rules’ did not always extend to the close regulation of hours. This could lead to lost wages for the workers and fines for the mill-owners if the inspectors took action. More than a decade after Webb's analysis, the secretary of the Oldham Master Cotton Spinners' Association wrote to the GPO:

It was an ambitious plan, with the association proposing to supply its two hundred members in mills scattered across a wide area with time from their master clock ‘receiving Greenwich time from the Post Office’.Footnote ⁴⁴ A cynic might find it interesting to note that this employers' association proposed to distribute its own time from a master clock in its own office, that clock alone corrected by the Greenwich time signal: ‘The Association are … desirous that the signal should be given at the Association's offices, and transmitted thence to the various members of the Association.’Footnote ⁴⁵ In other words, every mill in the association would get a clock telling the same time as all the others, but which could, in theory, not tell actual Greenwich time, if for example the synchronizing signal failed to correct the clock. This was the nature of electric master-and-slave clock systems: they provided a standard time but not necessarily Greenwich time.

The GPO was none too happy with Oldham's proposal, but its concern was the loss of possible revenue. GPO practice was to prevent local area networks being erected by private concerns and to provide individual clocks with their own synchronizing signal – in this case, a prospective additional two hundred clients, representing a significant amount of tariff income, especially as some of the mills were several miles from the nearest telephone exchange or post office.Footnote ⁴⁶ In July 1914 GPO officials met the Oldham association's secretary, with the local factory inspector present, to explain their position. But the association was ‘not willing to vary their proposal’, citing the presence of a further time-distribution company operating in London, the Greenwich Time Company (a shorter-lived competitor to STC), and requesting ‘to be placed in a position to do for its members what the Greenwich Time, Ltd, offer in London’.Footnote ⁴⁷

Again, the GPO's response was formulated to protect its own revenue-earning capacity, not to prevent the potential degradation of quality of the time signal received at each mill: ‘It is not considered advisable that the PMG [Postmaster General] should accede to the request … which will establish in the provinces a type of business actually in competition with the GPO itself.’Footnote ⁴⁸ In any case, so much time had passed by this point that the problem had gone away, at least temporarily: ‘In consequence of the present condition of trade, under which many of the mills are closed or working less than ordinary hours, due to the War, it is thought the question will be held over for some time.’Footnote ⁴⁹

If factory legislation were largely mute on the subject of defining time, we can find more explicit evidence for its significance by exploring that other great Victorian social concern: alcohol. The regulation of the sale of liquor proved to be one of the most controversial pieces of proscriptive legislation of the nineteenth century.Footnote ⁵⁰ The sheer volume of legislative paperwork – thirty-one statutes from 1828 to 1889Footnote ⁵¹ – bears vivid testament to the controversy. A. E. Dingle comments that ‘Victorians were obsessed with alcohol. Among the literate and the articulate, the proper place of drink in society was debated with an intensity and an exhaustiveness which is now difficult for us to comprehend’.Footnote ⁵² Part of this obsession with the control of the demon drink could be expressed through temporal limits. The 1872 Licensing Act is commonly held to be the first that introduced nationwide restrictions on the hours of sale of liquor to the public, and the discussions over just this act alone, never mind all those that preceded and followed it, were long and acrimonious.Footnote ⁵³ In the sale of alcoholic drink, issues of class, freedom, public health, morality and the role of the state in circumscribing legal human activity came together in legislation that is a perfect exemplar of nineteenth-century standardization and regulation, and time was the tool by which it could be policed. For the Victorians, it seems, temperance embodied temporality.

The 1872 act, and its swift successor in 1874, provided limits on hours of liquor sale, but like the Factory Acts already discussed it did not define whose time would be closing time. It would be wrong to assume, therefore, that Greenwich time had already become embedded as the time for all Britain. Evidence against this assumption comes in 1874 in the context of parliamentary debate about Richard Assheton Cross's intoxicating-liquors bill. The Times reported the outcome of a survey of local authorities to determine how the 1872 Licensing Act was working. The results included examples of local public opinion about the restrictions. The Times was clearly very much in favour of the act and against the changes being proposed in Parliament at the time. It noted,

Yet it was serious. It was not to do with the difference between the two times but a definition of one over the other, and, moreover, the ability of both publican and inspector to get access to the same (trustworthy) time. In Parliament a few days later the matter was raised again. Sir John Kennaway, a Conservative MP,Footnote ⁵⁵

In other words, time applied to regulation was not about the actual time of day but about one single time being used and agreed on. Kennaway's proposal fell on deaf ears:

Whilst Kennaway was confident that publicans could get access to Greenwich time from their local railway station, in fact Goldsmid's response was unwittingly closer to the mark. Many public houses in London, in the space of just a few years, were indeed connected with the Greenwich Observatory via STC's commercial service, albeit indirectly. Kennaway had been onto something important here. In a critical review of the 1872 and 1874 Intoxicating Liquor Licensing Acts, James Paterson, a prominent barrister, observed that before 1880 ‘the justices were not bound to follow Greenwich time, but followed the medium [i.e. mean] time of the place, as in Curtis v. Marsh [sic]’.Footnote ⁵⁷ This legal case of 1858 may have been one of the ‘several reasons’ that Cross had to support Kennaway's clarifying amendment. That year, in Dorchester, the case of Curtis v. March decided that court time was to be local mean time, not Greenwich time.Footnote ⁵⁸

In later life Cross described his 1874 bill as ‘not much to be proud of’, but he went on to become a significant social reformer during his period as Home Secretary.Footnote ⁵⁹ His last proposal, in 1880, was to regulate all London water suppliers under a single authority much as the telegraphs had been brought together ten years before, but he only just missed out on being present when his wish was finally granted and the issue of a standard legal time for the UK was resolved. In August 1880, just a few months after Cross left Parliament, six years after his discussion of time standardization and distribution in liquor licensing, the Statutes (Definition of Time) Act was passed, defining Greenwich Mean Time (GMT) as the definitive time standard that would apply to all legislation.Footnote ⁶⁰

The Royal Observatory had been actively exporting GMT since the 1830s, with its 1833 Time Ball and from 1836 the ‘hand-carried’ commercial time service provided by members of the Belville family who carried around a chronometer set to time at Greenwich.Footnote ⁶¹ As we have seen, since 1852 the Observatory had provided GMT across telecommunications networks. The 1880 act was a piece of legislation ‘to remove doubts as to the meaning of Expressions relative to Time occurring in Acts of Parliament, deeds and other legal instruments’. The extent of the act (which is about as concise as possible, and was passed through Parliament without discussion) was to state that

At last the law provided a definition of time, twenty-two years after Curtis v. March and six years after the liquor debate raised the question of whose time is closing time. The passing of the 1880 Time Act established formally that legal time meant Greenwich time. In a lecture to telegraph engineers in 1881, the architect and inventor of the STC system reinforced the point that his work was about standardization. He had invented ‘a system for ensuring uniformity of time between our public and private clocks’.Footnote ⁶³ There is an assumption in Derek Howse's standard work and elsewhere that there was a rapid and inevitable de facto standardization initiated in 1852 and completed long before the 1880 Time Act, the latter therefore being merely a catching-up instrument. Greenwich time, so the account goes, was a trustworthy and unambiguous standard, easy to access and technologically straightforward, and (despite the usual Luddite grumbling about ‘railway aggression’) there was plenty of demand for it, so it proliferated, pretty much straightaway. Yet the facts that Curtis v. March took place six years after the start of the Greenwich time service by telegraph, that the liquor licensing time debates occurred as late as the 1870s, that factories had problems getting access to GMT well into the twentieth century, and that a legal timescale was only defined in 1880 appear to reinforce the suggestion in this paper that the situation was not as clear-cut, or as rapid, or indeed as inevitable, as it might have seemed. Moreover, the received historical view that the technology was relatively straightforward also requires adjustment, given the evidence in the previous section.

Now that the standard was defined it remained to make it accessible. In the case of liquor licensing this meant large numbers of public houses needing to set their tavern clocks to Greenwich time. Notwithstanding Sir John Kennaway's confidence that they could do so by checking with the nearest railway station, it would appear that many pubs in London became in effect indirectly connected to the Greenwich Observatory. According to STC's subscriber list for 1886 a total of eighty public houses, hotels and restaurants in London took their hourly telegraph time service.Footnote ⁶⁴ This is a surprisingly large absolute figure even given the size of central London, but more astonishing still is the relative figure. Licensed premises comprised twenty-five per cent of the total number of subscribers taking STC's service in 1886 (total 313), a substantial proportion and one that would seem to bear out the proposal that liquor licensing legislation, whilst mute on the matter of timescale definition, nevertheless created a demand on the part of licensees for a uniform time standard distributed widely amongst licensed premises. Of course there will have been other reasons why any particular pub took STC's service. The Mail Coach in Farringdon Street, for example, happened to sit right next to the Sunday Times newspaper offices. This perhaps enabled newspaper workers to keep one eye on their deadlines while taking refreshment.Footnote ⁶⁵ But it must be assumed that many publicans wanted a clock on the wall telling trustworthy and standardized (Greenwich) time to which they could point when the inspectors came to visit. Their licence and livelihood were newly at stake if they got the time ‘wrong’. It may therefore be no coincidence that when STC's business faltered in 1896, it was bought from the receiver by one John Kershaw O'Donoghue, who also happened to be the general manager and secretary of the Licenses Insurance Corporation and Guarantee Fund, a company founded in 1891 to provide insurance to licence-holding publicans against loss of trade if their licence were to be revoked for some reason. Buying a company that linked his clients, the public houses, with Greenwich time, therefore decreasing the risk of licences being lost and of claims under policies, might possibly have appealed to O'Donoghue as a sound business proposition. This network of real clocks and wires can be viewed as the making portable of the Greenwich time standard, the electromagnetic clock correctors physically marking off Greenwich hours miles away from the Observatory itself. The legislation had changed the value of Greenwich time for a discrete set of people (publicans). As M. Norton Wise observed, ‘the relevant audience must agree that any proposed measure registers something valuable for them about quality, and not some artifact of arbitrary or interested opinion’. Liquor legislation made Greenwich time valuable and STC rendered it portable.Footnote ⁶⁶

Building networks of ‘power’ and ‘intelligence’

This story of a standardizing culture recognizes that the development of a time-distribution network comes partly from the demands of society at large, its concerns with alcohol and working conditions, and the maintenance of a chronometer industry servicing a powerful imperial navy, and partly from the possibilities afforded by technologies. As we have already met the social reformers, the legislators and the railwaymen, let us now meet the network-builders, the wiremen and the telegraph speculators keen to promote new opportunities and indeed new companies in the marketplace. It is notable that the privately owned STC, which commenced operations in 1876, figures so prominently in time distribution in this period against a backdrop of a state-owned monopoly which itself dates from 1870 with the creation of the GPO Telegraph Department. Here is evidence of the complex process of network-building and the long duration of ‘interpretational flexibility’ as new systems grow up. Upon examining STC's correspondence with the Telegraph Department in the 1880s, it is clear that London's public telegraph network was still very much a contested space and that policy as to how the network should be operated and policed was being developed on a case-law basis through a series of what amounted to gentleman's agreements. The telegraph system was provided by a monopoly company – the GPO – which was itself the regulator of that monopoly. Whilst this caused tensions in the GPO's relationship with related companies, it appears there was a desire amongst the Telegraph Department's senior staff that those tensions should be managed fairly, or, conceivably, that the private sector found ways of securing the goodwill of those staff. In part this may be because the GPO and therefore the state's ownership of the telegraph network was itself relatively new. The culture was the same: all sorts of businessmen and engineers had been associated with the technology, but after 1870 some found themselves in the state sector (i.e. the Telegraph Department) and some in the private sector (e.g. STC), so many of them would have been familiar with the entrepreneurship and empire-building that surrounded the technology of the telegraph. Thomas Hughes has described the parallel situation in the electric power industry as a ‘messy economic, political and social vitality’ embodying the ‘complex objectives of modern men and women’.Footnote ⁶⁷ The relationship between the GPO and STC exhibited just such ‘messy vitality’: the telegraph network had grown up organically, over time, based on commercially driven ad hoc agreements rather than being subject to an overall design, and attempts to bring it to order, to a standard form, to regulate it, were bound to be messy. This social, rather than scientific, approach to network-building means that an examination of the people involved in the story is crucial to an understanding of the eventual shape and characteristics of the network.

The uneasy entente between STC and the GPO was revealed when in 1882–3 the STC's predecessor, the chronometer-makers Barraud & Lund, transferred their telegraphed time-synchronization business to a company soon to be called the Standard Time Company. It started with an innocent enough request, that the GPO engineers transfer the telegraph wires from Barraud & Lund's in Cornhill, City of London, to the offices of the new firm, a few hundred metres away in Queen Victoria Street. The detail of the ensuing correspondence is less important than the fact that it opened a veritable can of worms, revolving around what was alleged to be STC's onward operation of a private telegraph subscriber network on its own wires connected to those of the GPO. A statement setting out the networking arrangements between the GPO and STC was prepared by GPO staff, who told the Telegraph Department secretary that the

In pencil beside this note the secretary tellingly scribbled, ‘This they had no business to do.’ On 18 May 1883 the GPO wrote back to the company, outlining proposals for the terms and conditions of a new agreement. The secretary berated STC for its activities leading up to the proposed transfer to Queen Victoria Street:

STC promptly deployed evasive tactics:

Not only was there a gentleman's agreement to work over each other's lines, STC argued, but the GPO had been complicit in the technical protocols STC had followed in carrying it out:

Yet it seems that the technical quality of STC's own line-work was not what concerned the GPO. Rather, it was the matter of lost revenue. In an internal memorandum following receipt of the letter from STC, it was noted that

The GPO's written response endorsed this notion of a clean sweep, putting aside older, more collaborative, practices to bring the arrangements into a standard regulatable form:

So as to avoid going deeply into the further correspondence, suffice it to say that a certain amount of horse-trading ensued and compromises were reached, but the situation remained essentially unstable and unresolved.

In this context it is possible to read the development of the London telegraph system and the GPO's relationship with STC as an attempt to regulate a heterogeneous and unruly situation, to bring into a standard form a homogenized state-controlled network emerging from the wider nineteenth-century move to regulate and to standardize, a theme widely discussed in cultural writings in the history of technology. We have already seen an episode in which the GPO, which operated a relatively new telecommunications monopoly, sought to regulate fringe telegraph operations such as time signals, which were largely outside its own field of business, by mediating a tense relationship between private firms and its own engineers. It had one eye on its own revenue and another on the setting of precedents that would be beneficial to its own business in the future. In the case of STC's operations in London, a compromise of sorts was fashioned in which unruly and irregular activity had a line drawn under it, with future activity falling under new, regularized and standardized agreements. But in the background the issue was subjected to close scrutiny with a view to establishing a legal opinion (on what was almost a philosophical question) that could bolster the regulator's position.

The bickering between the GPO and STC in 1884, following its attempt to move to new premises, and the resulting arguments over alleged infringements into Telegraph Department territory, rumbled on for a few months. Lists were prepared and accounts settled. Matters might have been left to lie were it not for STC's stock market flotation in 1886, an abridged prospectus for which was published in The Times and the Daily Telegraph, listing the company's subscribers. It included the paragraph ‘A contract has been entered into with the Metropolitan Railway Company for the synchronising of the clocks throughout their entire system.’Footnote ⁷³ But one of the Telegraph Department's senior staff picked up on a small detail that only appeared in the full prospectus and promptly blew the whistle on the entry, which stated, ‘The synchronizing wires can be run underground as well as overhead, and the cable through the whole of the Metropolitan Railway gives immediate access to the most populous parts of the metropolis.’Footnote ⁷⁴ The Telegraph Department secretary was asked, ‘Is anything known as to this cable and its relation to the Postmaster General's exclusive wayleave over the Company's system?’Footnote ⁷⁵ The chief engineer's department was in the dark: ‘I did not know of the existence of this cable, did you? If not can any of your people find out anything about it. Does it do more than synchronise the Metropolitan Railway Company's clocks? Is it carried beyond their line?’Footnote ⁷⁶

The Postmaster General enjoyed ‘the exclusive privilege of transmitting telegrams within the United Kingdom’, whereby ‘the term “telegram” means any message or other communication transmitted or intended for transmission by telegraph’. Exceptions included private telegraphs within businesses, relating solely to that firm's work, but not between two different merchants.Footnote ⁷⁷ The GPO's initial observations suggested that the Metropolitan Railway Company (MRC) ran its own wire along its own railway throughout, connected to an STC wire at Aldgate Station which synchronized its clocks along the way to Notting Hill Gate Station, except for one wire loop which left the railway at Baker Street to serve the adjoining Globe Hotel.Footnote ⁷⁸

Were it not for that one loop, the railway would easily have been able to argue that its was a wholly internal telegraph and therefore fell outside the privileges of the GPO. But the loop existed and the chief engineer's recommendation to the secretary was that ‘in this instance the legal powers of the Railway Co. have been exceeded’. On legal advice the secretary wrote to the MRC: ‘the erection of the cable on the Railway would seem to be in contravention of the Postmaster General's exclusive rights of wayleave’.Footnote ⁷⁹ The MRC replied robustly:

He went on,

The GPO solicitor's view on this response cut to the heart of the matter: ‘There seems in fact to be no doubt that the wires in question are telegraphs, the only question is whether the setting of a clock constitutes a message or other communication.’Footnote ⁸¹ The GPO's response to the railway company was a masterpiece of legal fudge:

The MRC wrote back to say that ‘such a view of the Post Master's right of interference is not recognized by this Company’. Its legal advice had been clear:

The matter was put aside until the following year, when further legal advice sought by the GPO reinforced the MRC's own opinion: ‘it is quite clear that no action should be taken in the case of the Standard [Time] Company’.Footnote ⁸⁴

Counsel's extensive consideration of ‘whether the performance of the synchronising operation by means of an electric current passing through a wire is the transmission of a message or other communication from the Company to their subscriber within the meaning of the Telegraph Acts’ touched on fundamental issues of networks and of standardization. One of the conclusions stands out: ‘I think the idea of telegraphic communication necessarily involves the existence of an intelligent operator, by whom the communication is sent.’ Following the connection of a subscriber's clock to STC's standard clock, and the synchronization of the former to the latter each hour,

With this useful summary of the contested nature of time standards and the implied role of certification and trust in a successful standard-distribution network, it seemed that the matter could come to rest. But not so. A quarter of a century later the issue was raised again, in the early years of the Great War, following a string of applications to the GPO from provincial customers wanting to set up their own local telegraph networks, distributing time signals from a single master clock set to Greenwich time by a GPO time signal. With the intervening twenty-five years of technical development a clearer understanding of the different forms of electric timekeeping had emerged. The GPO engineer felt that ‘considerable developments which could not then have been foreseen have since taken place in this class of business … it may be well now to review the 1888 decision in the light of present knowledge and conditions’. He noted that ‘it would appear that the fundamental principle which it is desired to establish lies in the distinction between the transmission of power and the transmission of intelligence’.Footnote ⁸⁶ In the eyes of the engineer, in the latter case it did not matter whether the intelligence was conveyed from human or non-human media (for example the playing of a gramophone record through a telephone handset) – the result was still the transmission of ‘intelligence’, not merely ‘power’, and STC was therefore sending messages.

The legal–philosophical debate over the nature of a ‘message’ can be seen as a metaphor for debates of much wider contemporary concern, also involving power and intelligence. Ben Marsden and Crosbie Smith noted the idea of telegraphic ‘intelligence’: the (rhetorical) uses and advantages advertised by telegraph company promoters to garner support for their latest fundraising exercise. In their survey, different forms of telegraph might ‘send instructions and edicts, enhancing surveillance and centralised control’; ‘strengthen monarchy, entrap criminals and keep friends in touch’; ‘preserve public order, or economise steam engines’ use for the business entrepreneurs of the new railway system'. And so on.Footnote ⁸⁷ Iwan Rhys Morus has also examined the taut relationships between telegraphy and intelligence in the Victorian cultural landscape. He demonstrates that the telegraph allowed human intelligence to ‘proliferate and impose its discipline increasingly further away from its point of origin. The telegraph was easily seen as an instrument devoted to the spread of Victorian values’.Footnote ⁸⁸ As well as shrinking space the telegraph could telescope time: ‘The crucial feature of the telegraph to many of its early promoters was its seeming capacity to make intelligence instantaneous.’ Morus raises the issue of telegraph time signals and their capacity to hide from the public the actual day-to-day complex set of underlying operations, some of which have been exposed further in this paper. Speaking of the RO's time-signals project, Morus considers ‘the amount of labour required to maintain such an apparently effortless, seamless web. Embodying intelligence and its transfer in a network of machines required constant human intervention and regulation to maintain its façade of seamlessness’.Footnote ⁸⁹

Across the same period of STC's operation discussed here, ‘power’ (i.e. electricity and its transmission, initially for electric lighting) was very much a focus for debate and discussion. There were controversial and heavily contested Electric Lighting Acts in 1882 and 1888. Legislation relating to electricity was discussed and formulated throughout the period leading up to the Great War.Footnote ⁹⁰ This was a time of growth and development, with electric lighting and then motive power being introduced to London's streets in a blaze of popular rhetoric. Electrical engineers were likened to Michelangelo or the Colossus.Footnote ⁹¹ In a wider sense, though, discussions about power raised comparisons with imperial and state power, and the use of telegraphy (primarily submarine telegraphs, but the connection also stands here) to manage and propagate that power. ‘What is wanted in darkest Africa is the electric light’, thundered a graphic contemporary electric-lighting advertisement framed around explorer and journalist Henry Morton Stanley, and the public image of electrical networks clearly included a strong imperial dimension reflected back to the metropolis from the dark periphery.Footnote ⁹² Daniel Headrick has noted of the submarine cable that

In a minor but revealing way the power struggles between STC and the GPO may perhaps be symptoms of this wider Victorian conflict.

For all its significance, this specific episode of legal wrangling was to come to no particularly decisive conclusion. The distinction between power and intelligence was never pressed, as practical considerations eclipsed the discussion. The Telegraph Department secretary's final word on the matter, in 1916, was pragmatic:

Understanding the wider context of power and intelligence relating to the electric telegraph helps us interpret the discussion surrounding the nature of STC's time-signal service, providing a reading of the exchange at a deeper level than a superficial debate over light bulbs and human telegraph operators.