Carl H. Kraeling, in “The Episode of the Roman Standards at Jerusalem” (Harvard Theological Review 35.4 [Oct. 1942] 263-289, via Laudator Temporis Acti), provides just the sort of detailed philological analysis that I revel in:

The word σημαία used by Josephus in his account of the episode of the standards is, like its Latin equivalent signum, a generic term and may apply to any or all of the standards borne by military units, though it is used also in a narrower sense for one particular type. Among the Roman standards the first to be mentioned are the aquila, a golden eagle mounted on a pole, and the imago or imagines, representations of animals or busts of the Emperor similarly mounted.[13] Both types are essentially symbolic and religious in their significance. The aquila borne by the aquilifer is the palladium exclusively of the legion. Legions also have imagines borne by imaginiferi, but they share this type of standard with other troops, the urbaniciani, the vigiles, the alae and the auxiliarii. The theriomorphic imagines, comprising mainly zodiacal animals, have something to do with the dies natalis of the unit. The images of the Emperor, what ever else they may denote, have a religious and cultic significance also. While every established military unit could, and perhaps did, have its own theriomorphic imago, it is clear that some units did not have separate representations of the Emperor. What the criterion for the distribution of the imperial likenesses may be, is not yet entirely evident.

The next type of standard to be mentioned is that to which the word signum is applied in the narrower sense.[14] More familiar than the others if for no other reason than because of representations in the school texts of Caesar’s Gallic Wars, the signum consists of a spear decorated just below the spear-head with a cross-bar and fillets, and adorned along the shaft with a series of discs, or wreaths and discs, or wreaths and discs and mural crowns. So far as the discs (phalerae) are concerned the signa can be divided into two types, those that are aniconic and have smooth, polished surface, and those that are iconic, being embossed with a likeness of an emperor (or an image of a deity?). The signa, while also of religious significance, are basically the instruments of tactical procedure and hence essential to all troops engaged in tactical manoeuvres. Each military unit has as many signa as it has tactical elements, though in the case of a cohort the signum of the triarii maniple is simultaneously also the signum of the cohort as a whole.

The last type of standard to be mentioned is the vexillum, a cloth flag attached to a cross-bar hanging from the top of a pole or spear. It is used by temporary detachments from established military units, which are therefore known as vexillationes, and in cavalry alae. Under what conditions it served as an identifying medium and as a tactical instrument respectively, is not entirely clear.

It’s useful to be reminded that in Greek texts dealing with Roman topics a Greek word (like σημαία here) can be simply an equivalent of a Latin one, so that there’s no point trying to apply its usual range of Greek senses. For the footnotes, see the Laudator post. And if you’re wondering about maniple, it’s (OED) “A subdivision of the Roman legion made up of two centuries, numbering 120 or (for some purposes) 60 men”:

Etymology: < Middle French maniple […], Middle French, French manipule […] < classical Latin manipulus handful, bundle, sheaf, unit of infantry < mani- mani- comb. form + a second element < the base of plēre to fill, plēnus full (see pleni- comb. form).


  1. D J EDDYSHAW says

    Borrowed by Britons, the Latin signum ended up as Welsh swyn “magic charm, spell”, which doubtless proves … something.

  2. D J EDDYSHAW! Did I see you spin at the Amnesia in Ibiza? Seamless transitions, dude!

  3. Of course, in mathematics, signum is the name of the function sign (x) = x/|x| (or 0 for x = 0). In this case, I think the Latinized name was probably created just to distinguish sign (x) from the homophonous sin (x).

  4. Dan Milton says

    Your reminder of the pitfalls of Greek terms for Roman topics reminds me of a Jehovah’s Witness once trying to convince me how wrong the usual Christian cross is because stauros is simply a stake.

    (With Covid easing, will door-to-door missionaries get back in action? I always enjoy them.)

  5. J.W. Brewer says

    “Vexillation” feels like it ought to mean more or less the same thing as “vexation” (maybe crossed with “pixilation”?) rather than being a technical term for a sort of military unit.

  6. John Cowan says

    L vexillum ‘flag, banner’ is the diminutive of vēlum ‘curtain, cloth, awning’, also (usually pl.) ‘sail’, cognate with English wick. Vexare is ‘shake, jolt; harass, annoy; vex’, cognate with English quake. So they have different stem vowels.

  7. Tactical use basically means showing the specific troops where they should move. One of this cases is probably known to anyone who had enough Latin to read about military campaigns, i.e. signa inferre “to attack”, literally “carry the signs in(to the enemies ranks / enemy territory)”.

  8. David Eddyshaw says

    With Covid easing, will door-to-door missionaries get back in action? I always enjoy them.

    I don’t see why not. We had door-to-door canvassing back again in the local (and Welsh Senedd) elections last month. Same principle …

    Did I see you spin at the Amnesia in Ibiza?

    D J EDDYSHAW is my glamorous alter ego. Men want to be him, women want to be with him …*

    * OK, he’s a bit retro. At least I’m woker than he is. The bastard.

  9. Borrowed by Britons

    I have read that the dragon in pendragon originated in Latin dracō, literally “dragon”, but also the name of the standard used by Roman cavalry unit in the later Empire that had the form of a windsock with the head of a dragon or mythological beast.

    Middle Welsh pendragon, pendreic “warrior, hero, war leader, chieftain, prince, military power”, epithet applied to Uther and later to Arthur, is literally “chief of warriors, chief of chieftains” and a compound of pen “head” with dragon and dreic, “warrior, chieftain”, both ultimately from Latin dracō, “dragon, Roman cavalry standard in the form of a dragon windsock”. Here is some information on the dracō standard:

    However, the exact details of the semantic development in Brittonic have never been clear to me.

    Did the development of dracō to Middle Welsh dragon, dreic, “warrior” come about from the fact that the signifer called dracōnārius rode prominently, in the first rank, and there was generalization from “warrior” from there? Or that dracō standard symbolized the cavalry and thus by metonomy came to be used of a chieftain himself? Or is there some other scenario, with “warrior” being from “dragon”, directly? (I wonder if there are any parallels for this… 龍爭虎鬥 Enter the Dragon (Cantonese literally, lung4 zaang1 fu2 dau3 “Dragons Compete, Tigers Fight”, a fierce struggle between evenly matched opponents).)

  10. David Eddyshaw says

    I hadn’t come across draconarii before. Thanks!

    I like the idea that the Welsh dragon may be a Sarmatian. I expect she’s* an early Polish immigrant …

    * Draig is feminine, though this doesn’t actually have anything to do with sex. I think dreic is still feminine grammatically in the sense “warrior.”

  11. Stu Clayton says

    Is there any evidence that the Dragon Lady was a draig queen ?

  12. David Eddyshaw says

    I think we have all the evidence we need. It all fits, I tell you!

  13. Stu Clayton says

    It’s enough to give fitness* a bad name. In certain epistemologies, and in widespread practice, “internal consistency” is regarded as a necessary criterion of … well, I dunno. Truthiness maybe ? Usefulness, reliability (on what?) ? Advocates of this idea seem to be unaware of the existence of hard-core crazies with their conspiracy theories (for example).

    * Not fitness for a purpose, but fitness as the state in which puzzle pieces find themselves when the puzzle is completed. To distinguish this concept, I recommend the neutral German weasel word Paßförmigkeit.

  14. Lars Mathiesen says

    vexillum is the diminutive of vēlum — l – l dissimilating to x – l? Call the neogrammarian police! (I am not looking this up and risking a boring explanation).

  15. David L. Gold says

    Velum with diminutive vexillum is not isolated in Latin, which also has at least ala, palus, and paulus with diminutives axilla, paxillus, and pauxillus.

    That is no explanation of how those diminutives arose from their base forms but the four examples do show that randomness is not present here.

    The fact that the base forms have /l/ may be significant.

  16. The -x-’s are retentions from an older form of the word: āla < axis, velum < PIt. *wekslo-, etc.

  17. January First-of-May says

    …Oh, so *vexlum – vexillum > vēlum – vexillum (and ditto for the others). Probably with compensatory lengthening too That makes surprisingly much sense.

    IIRC there’s a very neat derivation in Russian that includes a lot of similar remnant-consonant cases, but offhand I can’t recall the exact example.

  18. PlasticPaddy says

    The [V]l- to [V]xill- correspondence seems to have been analogically productive, because there is also pālus and paxillus,

  19. David Eddyshaw says

    If Wiktionary is to be believed, pālus is also a case of *-ksl-, so it would be an actual parallel rather than a case of analogy.

  20. The word maniple immediately made me think of this (which I first discovered via this post.)

  21. pālus is also a case of *-ksl-, – for a Russian reader of T. Yansson with her Tofslan och Vifslan* derivations like māla < Proto-Italic *smakslā look suspicious.

    *Tillsammans har de utvecklat ett eget språk, som låter "ungefärsla så härsla om ni förstårsla"

  22. Om ni förstårsla.

  23. David Eddyshaw says

    derivations like māla < Proto-Italic *smakslā look suspicious

    We’re talking about a language in which *-sr- has become -br-. No phonetic perversion was beyond those ancient Romans.

    (The French have carried on the grand Roman tradition: ad illum -> /o/, falling together with aqua.)

  24. “In hoc signo vinces”?

Speak Your Mind