SPECIAL

CAPTION THIS CAPTION THIS CAPTION THIS CAPTION THIS

CAPTION THIS CAPTION THIS CAPTION THIS CAPTION THIS
CAPTION THIS CAPTION THIS CAPTION THIS CAPTION THIS
© Angel García Pinto

44 Ancient Warfare XIV-3

 context of warfare. In his Parallel Lives, we find various Spar-
SECRET COMMUNICATION IN ANTIQUITY tans that successfully sent and received secret scytale-messag-

The Spartan
es in the context of warfare, including the Spartan command-
ers Lysander, Clearchus, and Agesilaus.
Two complete descriptions of the use of the scytale as a cryp-

scytale
tographic device can be found, in Plutarch (Lysander, 19.5–7) and
in Aulus Gellius (Attic Nights, 17.9.6–16). Both authors were ac-
tive in the second century AD so were writing at some consider-
able temporal distance from the period in which the scytale were
used: in the fifth and fourth centuries BC, and especially the pe-
Communication security is of major importance to our riod between the outbreak of the Peloponnesian War in 431 and
the Battle of Leuctra in 371. Nevertheless, Plutarch and Gellius of-
modern world. The gathering of intelligence and spying
fer useful accounts of the scytale that can tell us much about how
on one’s enemies is essential for any government in this deceptively simple cryptographic device could and would
order to determine the political and military direction have been useful to send encoded messages in the field.

of the state, especially in times of conflict when es- Sending secret messages
sential information on enemies can obviously facilitate The sending of secret scytale-messages worked as follows, ac-
the war effort. Since antiquity, individuals in all civili- cording to Plutarch (Lysander 19.5–6):

sations have been trying to encipher confidential cor- When the ephors send out an admiral or a general, they
make two round pieces of wood exactly alike in length and
respondence – mainly in a military context – while oth-
thickness… and keep one themselves, while they give the
ers have been trying to decipher it. The ancient Greeks other to their envoy. These pieces of wood they call scytalae.
and Romans in particular seem to have been experts Whenever… they wish to send some secret and important
message, they make a scroll of parchment long and narrow,
in the art of secret writing, known as cryptography. like a leathern strap, and wind it round their scytale… After
doing this, they write what they wish on the parchment…
By Martine Diepenbroek and when they have written their message, they take the
parchment off, and send it… to the commander.

O
ne such example of ancient cryptography we
find in fifth and fourth century BC Sparta. Ac- Aulus Gellius’ description later in the second century is very
cording to these sources, the ephors (leaders) similar to Plutarch’s (Attic Nights 17.9.6–16):
of Sparta used a scytale – literally ‘a stick’ – to [When the] ancient Lacedaemonians… wanted to conceal
communicate with commanders in the field and disguise… public dispatches sent to their generals…
on confidential matters. This article discusses what exactly a Spar- [they] used to send letters written in the following manner.
tan scytale was and how was it used to create secret messages. There were two thin, cylindrical wands of the same thick-
ness and length… One of these was given to the general
The Spartan scytale when he went to war, the other the magistrates kept at
Greek sources vary widely in their discussions of the scytale home… [Then] they bound about the staff a thong… [and]
and, although they are broadly consistent in associating the they wrote the dispatch on that thong… When the letter
scytale with messaging, they do not always associate it with had been written… the thong was unrolled from the wand
secret messaging. According to our sources, the scytale could and sent to the general… This kind of letter the Lacedae-
and would have been used for a variety of purposes, including monians called skutalè [scytale].
as a messenger authentication stick, as a sort of military name
tag, and perhaps even as a reference to an erect penis/phal- In other words, before an admiral or a general was sent to war,
lus in a comedy. Another purpose of the scytale was its use he was given a scytale to communicate with the ephors in Spar-
as a cryptographic device. Following these sources, modern ta. The ephors and commander, the two communicating par-
cryptographers often do not see the scytale as a useful cryp- ties, therefore both had a scytale of the exact same length and
tography device. Instead, they often see it as a ‘toy cipher’ thickness. It was of the utmost importance that the two scytalae
that children could use to communicate with each other. From were the exact same size. Otherwise, the method would not have
Plutarch, however, we know that scytalae were highly likely to worked well, as will be discussed below. Whenever either party
have been used successfully as cryptographic devices in the wanted to send an important confidential message to the other,

Ancient Warfare XIV-3

45
they would take a thin strip of leather parchment and wind it antiquity, so there seems to be no particular significance attached
round their scytale, making sure that every bit of the stick was to the use of either as the medium for the scytale’s encrypted
covered in parchment. Plutarch used the word biblion to de- messages. It seems plausible that leather parchment would offer
scribe this writing material and its text, which may be translated a more robust material for messages that might need to be sent
in this context as ‘a strip of papyrus’ – adding that it was “like a over long distances, over difficult terrain, and in conflict situa-
long and narrow leather strap”, as if it were parchment (Lysand- tions since, according to both Plutarch and Gellius, the ephors in
er, 19.5). Aulus Gellius also believed it to have been a strip of Sparta used the scytale method to communicate with command-
parchment. He used the Latin word lorum to describe the writing ers in the field. Yet while on campaign one obviously had to be
material (Attic Nights 17.9.9). This can be translated as ‘leather’ able to write and send messages simply and quickly, so where
or, in this context, ‘a thin strip of leather parchment’. After wrap- papyrus was readily to hand this would presumably have offered
ping the strip around the scytale, the ephor or commander would a convenient alternative medium.
write whatever they wished to communicate on the parchment.
When they had written the message, they would unwrap the strip The first transposition cipher
from the scytale and only send the strip to the intended recipient. After a message was written on the strip of parchment or
This person only could then understand the intended message papyrus, the strip was then unwrapped from the scytale (Plu-
if he rewrapped the strip around his own matching scytale. For tarch, Lysander 19.5; Gellius Attic Nights 19.7.6–10). By
the method to work well, both scytalae had to be the exact same unwrapping the text strip from the scytale all letters in the
size and diameter and the strip had to be rewrapped around the original message were transposed to a different position, as
second scytale just as loosely or as tightly as it had been done the Plutarch and Gellius described. Thus, according to Plutarch,
first time. Otherwise, the letters would not have returned to their when a general in the field received a scytale message, he
original place. Although neither Plutarch nor Gellius mention it, could not have gotten any meaning out of it, since the let-
the secret message created by using the scytale could have been ters would have been disarranged (Plutarch Lysander 19.7).
written in either normal non-encrypted text – known as ‘plain- Gellius additionally mentioned that in a scytale-message one
text’ in cryptography – or as encrypted text – known as ‘cipher- had to deal with many partial and broken letters. The unroll-
text’ in cryptography. The latter option would have made the ing of the strip would have made the letters imperfect and
message doubly secret and, therefore, doubly secure. However, broken, and their parts and strokes were divided and separat-
nothing in their descriptions proves that the text was written in ed. This is because the complete scytale was covered in writ-
cipher. It will, therefore, be presumed that the scytale messages ing material, making the edges of the strip overlap (Plutarch,
were written in plaintext. Lysander 19.6; see example below). This changing or ‘trans-
posing’ of letters is known as transposition in cryptography.
Writing materials for scytale-messages The transposing makes the scytale the first transposition ci-
Since the sixteenth century there has been discussion among pher known in history, at least in theory, since although it is
scholars about what Plutarch and Gellius meant in their descrip- likely that scytale were used for secret communication in the
tions of the scytale. The Dutch scholar Erasmus, in his Adages, way described here, there is not clear evidence for this.
believed that both the stick and a leather thong wrapped around
the stick together were called a scytale. Erasmus potentially con- An example of a scytale-message
fused Plutarch’s and Gellius’ descriptions of the scytale here The following example shows how the transposition encryp-
since Plutarch suggested that either the stick or the leather strap tion achieved by this wrapping and unwrapping would have
(or, in fact, the message) were both called scytale, while Gel- worked, according to Plutarch’s and Gellius’ descriptions, and
lius mentioned that the letter that was sent to the general was how the strip of writing material might have looked during the
called a scytale. Cardano, in his 1550 work De Subtilitate, sim- subsequent steps in the process. For this example, the text “En-
ply mentioned cylinders (scytalae) that were used for secret com- emy attacks at dawn tomorrow” will be used. All illustrations
munication by the Spartans without giving any details on how are the author’s own.
this worked and what materials were used for it – something that First, a strip of writing material is wrapped about the
may indicate that Cardano knew about Plutarch’s and Gellius’ scytale. In this way small rectangular columns are created.
descriptions but found it unnecessary to repeat them in his own Since the wrapping is done by hand, it is likely that the edges
work (De Subtilitate 17.1036). Perrin translates the Greek biblion of the strip overlapped each other. Therefore, the columns are
as ‘parchment’ in the Loeb version of Plutarch’s Lysander, as Ed- not all exactly the same size.
gar Allan Poe had already done in his 1841 work A Few Words
on Secret Writing. And modern cryptographers seem certain that
parchment was used instead of papyrus. However, both papyrus
and leather parchment were typically used as writing materials in
Figure 1: Scytale with strip of writing material wrapped about it.

46 Ancient Warfare XIV-3

Then the text would have been written on the strip of writing
material. In Figures 2 and 3 (below), the text is written from ENEMY ATTACKS AT DAWN TOM
left to right and top to bottom on two lines. ORROW
Figure 5: Complete strip of writing material cut into pieces to show partial
E N E M Y A T T A C K S A T and complete letters.
D A W N T O M O R R O W We can see from this reconstruction just how challenging it
Figure 2: Text: 'Enemy Attacks at Dawn Tomorrow', written on scytale. would have been to attempt to reconstruct the original text
As Plutarch described in his account, once the strip of writ- from this scrambled ciphertext. It would certainly not be im-
ing material has been unwrapped from the scytale, all letters possible, but – without a scytale rod of the same size as the
will have been rearranged (Lysander, 19.7). By unwrapping one used in the original encryption – it would have been
the strip, the letters now appear per column, instead of per time-consuming. A good modern parallel might be the paper
row. So, instead of reading E-N-E-M-Y, the first word of the shredder, which offers one way of ‘encrypting’ or scrambling
message, one now reads E-D-N-A-E et cetera (see Figure 3). a confidential source text. With patience, skill, and time, the
In Figure 3, every letter of the message has been written on original text can be reconstructed from the shredded strips of
a complete piece of material strip. In other words, no letters paper. The advantage of the scytale device, however, is that it
were written over the edges of the strip, meaning that when offers the opportunity for that reconstruction to be managed
the strip was subsequently removed from the scytale stick, all much more quickly and easily.
the letters would have remained intact and would simply have
been rearranged in their order of sequence. Conclusion:
Greek sources vary in their discussions of the scytale and only
sometimes associate scytalae with secret messaging. The clear-
ED NA EW MN YT AO TM TO AR CR KO SW A T
est descriptions of the scytale as a cryptographic device come
Figure 3: Strip of writing material with text unwrapped from scytale. from Plutarch and Aulus Gellius. Although Plutarch’s and
However, according to Plutarch, the complete scytale was Gellius’ descriptions alone do not prove that scytalae were
covered in the writing material. This makes it more plausible used in practice for secret communication, the ingenious rear-
that the sender of a message did not in practice write with all rangement of the letters that they describe as the key feature of
letters neatly contained upon complete pieces of the scytale’s scytale communication makes the scytale a candidate for the
material strip, but that the letters of the message would also earliest known theoretical transposition cipher in history. Tak-
have been written across the edges of the strip (Figure 4). ing together all the extant sources on the scytale, it is wholly
plausible that Spartan scytalae were actually used for secret
communication – among other purposes – even though the
ENEMY ATTACKS AT DAWN TOM concrete evidence for this use remains largely inaccessible.
ORROW Indeed, the ingenuity and importance of the transposition sys-
Figure 4: Text ‘Enemy Attacks at Dawn Tomorrow’ written across complete tem of the scytale cannot and should not be overlooked here.
strip of writing material. In fact, the basic principle of the scytale – the transposing of
Since Plutarch’s description indicates the likelihood that the letters – can still be found in modern transposition ciphers that
whole strip was used for writing, at least some of the letters on are in use up to the present day. 0
the strip would have been written whole and on a complete
piece of strip. These letters would have remained intact and biobiobiobiobio biobiobiobiobio biobiobiobiobio biobiobio-
recognisable once the strip was unwrapped from the scytale. biobio biobiobiobiobio biobiobiobiobio biobiobiobio
Yet, since some of the letters on the strip would have been
written overlapping the edges of the strip, these letters would FURTHER READING
not have remained intact once the strip was unwrapped from ♦ Kelly, T., “The Spartan Scytale”, in J. W. Eadie &
the scytale. And indeed, according to Aulus Gellius, this is J. Ober (eds.), The Craft of the Ancient Historian:
what happened in practice: the unrolling of the strip made the Essays in Honor of Chester G. Starr (Lanham:
letters imperfect and broken (see Figure 5; Aulus Gellius, Attic University Press of America, 1985), pp. 141–169.
Nights 17.9.12–14).
If the strip from Figure 3 – for example – was unwrapped ♦ Kelly, T., “The Myth of the Skytale”, Cryptologia
from the scytale, the strip would look like the following figure 22 (1998), pp. 244–260.
(Figure 5). In this figure the strip is cut into pieces to show the ♦ West, S., “Archilochus’ Message-Stick”, Classical
difference between the partial and complete letters. Quarterly 38 (1998), pp. 42–48.

Ancient Warfare XIV-3

47