Marryat signal codes and their possible
application as ship identifiers in databases
Introduction
In 1817 Captain Frederick Marryat
introduced a system of flag signal codes for use by the merchant marine along
similar lines to the systems in use by the navies of the world. This system
included unique numerical identification codes for ships. There were competing
signalling systems but it is apparent that the Marryat system was the most
widely used by merchant ships before government sponsored identification
systems were introduced mid-century.
No systems of government-sponsored
signal codes for ship identification were in use by any nation during the first
half of the 19th century, as far as I have been able to establish.
This item investigates the Marryat system in some depth though with the
underlying objective of assessing the potential for using the ship codes as
numerical identifiers in databases for ships of this period. Illustrations of the flags themselves can be
found on a number of sites by google or yahoo searches.
The short answer to the question of the utility of the codes
for modern database purposes is “probably not” unless there exist other
documents that I haven’t found yet which economically and reliably link the
Marryat codes to additional information about the ships named in the code
manuals. However, I have documented what I have found about the Marryat system
in the manuals available to me as I have found little else of substance about
it elsewhere on the internet or in any publication other than the manuals
themselves. Copies of the original manuals, now mostly more than 130 years old,
are few and far between.
Update
Since writing this item I have
become aware of two publications by Sam Davidson which contain
additional information about Marryat’s code and a key to the ship names
compiled from multiple editions. If you are trying to identify a ship in a
painting of the relevant period these may well be easier to access than the
original manuals. The references are:
A.
S. Davidson, Marine Art And Liverpool, Painters,
Places, and Flag Codes 1760-1960, Albrighton,
Wolverhampton : Waine Research, 1986, pages 140 and. 144-167.
A. S. Davidson, Marine Art & Ulster: A Chronicle of Sail, Steam & Flag Codes. Upton, Wirral: Jones-Sands Publishing, 2005
Publication of the Marryat codes
The
original title was “A Code of Signals for
the Use of Vessels Employed in the Merchant Service”. The Library at the
From 1854 the title becomes “The Universal Code of Signals for the
Mercantile Marine of All Nations”. The Library at the
The full title of the 1861 edition
is "The Universal Code of Signals
for the Mercantile Marine of All Nations by the late Captain Marryat, R.N.,
with a selection of sentences adapted for convoys and systems of geometrical,
night and fog signals'' by G. B. Richardson, London, 1861.
I have had direct access only to the
1861 and 1866 editions of the code books. However, these contain editorial
information about earlier editions and are sufficient to make a general
assessment of possible research applications.
The 1861 Edition
The 1861 edition reprints the
preface to the Edition of 1856 which states that the 1841 edition was the last
to be edited by the original inventor. It gives the reason that the title was
varied to the "Universal Code of
Signals for all Nations" (in 1854) as being because of its “extensive
use on the continent of
Section I of the1861 manual lists British
warships with numerical codes up to 1467 and limited information about the
number of guns and type of ship.
Section II which had covered French
warships in earlier editions is shown as blank because the List of the French
Navy was no longer published.
The main section (III) covers
merchant ships. Sections IV‑VI list codes for lighthouses, ports and
operational signals. Unfortunately, the only information provided for merchant
ships is the Marryat code and the ship's name ‑ you do not even get the
tonnage, port of registration or means of propulsion let alone the nationality.
The inclusion of such additional information in the early issues of the Mercantile Navy List and the later
signal code lists in appendices to Lloyd’s
Register is critical to being able to use the later government allocated
codes for research data management purposes.
Evidently, the Marryat manuals were
always intended to be used in conjunction with other information. Shipboard
observers, if they could read the Marryat flags at all, would also be looking
at a national flag, it would be self‑evident whether the ship was sail or
steam and a professional ship's officer would be able to make a reasonable
guess at a ship's tonnage. As a manual for use at sea, the code book would
therefore be adequate but it has limited use as a historical record in the
absence of a means of efficient reliable linkage to other information. It seems
unlikely that pre‑1861 editions would contain more ship information than
later editions but it is not entirely impossible and should be checked.
The 1861 Marryat code signals for
individual merchant ships consist of a distinguishing pendant using the
numerals 1 to 3 followed by a number of up to four digits. If useful to do so,
these could be combined to derive 5‑digit numbers that could be used for
database identification purposes.
The 1861 numerical codes to be used
in conjunction with the first distinguishing pendant reach to 9876, those with
the second pendant also to 9876 and those with the third to 7640. However, some
numbers were omitted so the total number of ships listed is less than the sum
of these numbers. Numbers that would contain the same numeral twice in the same
hoist were omitted in the interests of minimising opportunities for error; thus
11, 22 ..., 111, 222 etc were omitted as were 101, 111, 112‑119, etc.
The 1856 preface made the offer to
publish ships’ official numbers as well upon request but the 1861 volume
includes only a few dozen ships’ details, apparently sent in by the owners, in
just two pages at the end of Section III. For these ships, the manual indicates
the British official number, tonnage, port and whether or not it is a steam
ship alongside each ship's name - but not
its Marryat code so that even the additional documentation in the manual itself
does not provide the information necessary to make the link automatically. The
first issue of the Mercantile Navy List (which
includes some more detailed information about every registered British ship
alongside its numerical and flag signal identifiers) did not come out until
slightly later than the preface to the 1856 volume of Marryat was presumably
written, so the publisher who possibly could have been aware of what was
envisaged for the MNL, may have hoped
to stake a claim for that market for himself. However, it is apparent from the
1857 MNL that the Commercial Code of Signals developed by
the British Board of Trade had already been so thoroughly established as to
render the Marryat system superfluous for British ships.
Changes in the 1866 edition
The 1866 edition introduced a fourth
distinguishing pennant. That made possible the allocation of codes to several
thousand more ships but the highest number allocated to be used with the 4th pendant
in the 1866 edition is 287. 1869 is the last edition that I have seen referred
to so the 1866 edition gives a quite late indication of the degree of adoption
of Marryat.
The 1866 edition also extends the
first, second and third pendant series by an additional range from 01 to 0987
(as distinct from 1 to 987). This would complicate converting Marryat codes
into 5 digit numbers by adding the pendant numbers as there would be two codes
logically translated as 10001, 10002 ... 10987. However, the technicality could
be easily accommodated by inventing nominal 5th, 6th and 7th pendants for
coding purposes to distinguish the 01…0987 sequences from 1 …. 987).
The 4th pendant listings indicate
clearly the principle of the deliberate omissions of some numerals to avoid
possible confusion that could arise from using the same number twice in the
same hoist. The omissions are 11, 22, 33, 44, 55, 66, 77, 88, 99, 100, 101, 110‑119,
121, 122, 131, 133, 141, 144, 151, 155, 161, 166, 171, 177, 181, 188, 191, 199,
200, 211, 212, 220‑229, 232, 233, 242, 244, 288…. etc. This information
is sufficient to derive the exact number of Marryat codes operative in 1861 and
1866 but it is sufficient for general purposes that the number is somewhat less
than 30,000 at a time close to 1870 when Bureau Veritas identified around
65,000 merchant ships in the world (refer Numbers).
The coverage is therefore significant so that the Marryat codes would have
value as ship identifiers in historical records if they could be linked to
other identifying information about the ships.
I have not found anything specific written about whether original
Marryat codes were ever reissued in order to reuse the codes of defunct ships.
However, there must have been recycling of codes. The lowest numbers assigned
to the first pendant are more or less continuous and it is inconceivable that
so many ships allocated codes even a decade earlier would still be in existence
in 1866. The prefaces of earlier editions may refer to this.
The 1866 edition also introduces a
new section for yachts of which about seven hundred are listed together with
their tonnage and rig. This may possibly constitute a fairly complete list of
larger British yachts at this date in a place where one might not expect to
find it. In fact, the large yachts of the wealthy may well have constituted the
majority of British yachts of this period as yachts for the masses were many
years in the future. The yachts were allocated 4‑digit numbers commencing
with 0123 to be used in conjunction with the club burgee.
The potential research applications
of the 1866 edition for merchant ships are as for the 1861 edition. The volumes
do not even confirm that the ship still exists as the text explicitly states
that "many" probably no longer exist and that specifically indicated
names will be removed in future editions if no confirmation of existence is
received.
Chapman’s “Auckland
Almanack”
Chapman's Auckland Almanack is essentially a “city guide”, albeit a substantial
one at the upper end of the scale. The 1862 edition contains a table listing
This discovery raises the
possibility that similar tables may have been compiled for other countries or
regions. Discovery of such a composite record for any of the major maritime
nations would warrant a much more positive conclusion about the utility of the
Marryat codes as ship identifiers in historical research.
General observations
An international system of ship
identifiers applicable to the first half or middle third of the 19th
century would be so useful for historical record purposes that Marryat should
not be completely written off for this application until it has been definitely
established whether any fuller documentation exists relating Marryat codes to
other ship details for one or more of the principal maritime nations.
Additional information even for only some ships, still has application in
helping to differentiate the remainder from others of the same name and thereby
helping to establish the magnitude and complexity of outstanding tasks of
identification and recording.
For those ships to which an official
number was allocated there is no useful database application for any other
code, other than for cross-checking and elimination purposes. However, the
elimination of ships with more useful later codes from earlier listings such as
the Marryat will leave a residual listing of ships* representing some sort of record of
international foreign‑going ships in the earlier period to which
information from other sources may be compared, thereby providing some sort of
minimum check list and research target.
* Roughly speaking,
those that did not survive 1857 in the case of British ships, later years in
the case of other nations.
Summary and conclusions
During the second half of the 19th
century government-allocated signal codes and official numbers provide a
superior alternative to the Marryat system of codes for the purposes of
identifying ships in a systematic manner in historical records.
The Marryat signal codes were
evidently the most widely used for merchant shipping in the first half of the 19th
century. The increase in the number of digits in the numerical codes about 1840
is probably a clue to when Marryat codes were allocated to a significant proportion of international
shipping. By the 1860’s they were being allocated to a substantial proportion
of merchant ships worldwide, probably the majority of those engaged in
international trade. By then, however, the Marryat codes had been superseded
already in their usefulness for identifying British and American ships and were
on the point of being superseded for the leading European nations by
state-allocated signal codes.
However, the Marryat code manuals
that I have seen for 1861 and 1866 do not provide the information required to
link the codes to other additional information about a ship such as place and
date of construction or to readily identify the degree to which codes of
defunct ships were being reallocated. Such additional information is essential
for these codes to be of any practical use in managing historical records. Such
a listing does exist for one year for one of the smallest maritime nations on
earth (New Zealand in 1862) but similar listings for a number of years for some
of the major nations earlier than the 1860’s would be necessary for there to be
any hope of the Marryat codes providing a world-level surrogate for official
numbers in the first half of the 19th century.
Without such a listing, the Marryat
code manuals would still have possible utility as a checklist against which a
project building up databases of mid 19th century European ships
from primary records could assess its progress and coverage but would be
unlikely to contribute much more directly.
If it was found useful to use the
Marryat ship signal codes as database identifiers for the earlier part of the
19th century they are already in a numerical form that could be
easily adapted to 5-digit database identifiers. Alternatively, they could be
converted easily into alphabetic codes similar in form to the later signal
letter codes by converting numerals 0 to 9 to the first 10 letters of the
alphabet. The need to allow for the initial pendant would give five letters
rather than four which would effectively differentiate them from the later
4-letter codes. As the Marryat codes were not duplicated by different nationalities
they would not require an additional national identifier as do the later codes.
Information sought
Information about the contents of
the editions before 1861 and after 1866 would be useful where it differs from
the above. It would be useful to obtain information about how many ships were
allocated Marryat codes, particularly during the 1840’s and earlier.
Particularly useful would be
references to sources for any lists of ships of any country that include
additional ship identifying information in conjunction with its Marryat code
and/or which match Marryat codes against the later Commercial Code of Signals'
codes or official numbers. Such information would be of much interest and have
many applications other than the narrow application investigated here.
Additional information about how
widely the Marryat codes were used later in the 19th century
alongside the later government-sponsored identification systems would be useful
to round out the coverage.
To go to “The allocation and use of ship identification
signal codes for merchant ships to WWI” click here
To go to “Signal codes as ship identifiers in
databases” click here
To go/return to main Maritime menu click here