The mechanism of Roman line relief

[Justin Swanton]

Now, "inter" has different meanings but we should not forget that first meaning is "between" rather than "inside, in the middle". From that (inter, intra) comes the Italian word "tra" which means "between".
"Inter ordines", "tra gli ordini" (literally) makes me think firstly to something happening between groups of people.

Inter does not have a primary meaning of 'between'. It is a rather general word, with the sense of 'among', 'between', 'in the midst of', 'amidst', 'during' (for time), etc., depending on the context.

When describing two objects inter means 'between': inter urbem ac Tiberim - 'between the city and the Tiber'. When describing a group or collectivity it means 'among', 'in the midst of': inter paucos - 'among the poor';  inter multitudinem - 'in the midst of the multitude'.

In the case of the ordines, we have a group of groups/collectivities: 15 ordines each composed of a number of legionaries. Does inter refer to the spaces between the individual units of the ordines - i.e. the legionaries - or the spaces between one ordo and the next? Answer: it could be referring to either. One needs to look at the context, in this case the context of which mechanism of line relief is better suited to the realities of combat, to deduce the more likely translation.

[Mark G]

There are other options still, Andrew.

Why, for example, must there be a solid front line at all?

[Patrick Waterson]

Now, "inter" has different meanings but we should not forget that first meaning is "between" rather than "inside, in the middle". From that (inter, intra) comes the Italian word "tra" which means "between".
"Inter ordines", "tra gli ordini" (literally) makes me think firstly to something happening between groups of people.

Inter does not have a primary meaning of 'between'. It is a rather general word, with the sense of 'among', 'between', 'in the midst of', 'amidst', 'during' (for time), etc., depending on the context.

I would also direct attention to the specific meaning of 'inter ordinium' in Lucius Junius Moderatus Columella's Res Rustica III.13.3, IV.14.2 and V.5.3, which is 'the space between two rows'.  See the Perseus lexicon entry here.

When describing two objects inter means 'between': inter urbem ac Tiberim - 'between the city and the Tiber'. When describing a group or collectivity it means 'among', 'in the midst of': inter paucos - 'among the poor';  inter multitudinem - 'in the midst of the multitude'.

In the case of the ordines, we have a group of groups/collectivities: 15 ordines each composed of a number of legionaries. Does inter refer to the spaces between the individual units of the ordines - i.e. the legionaries - or the spaces between one ordo and the next? Answer: it could be referring to either. One needs to look at the context, in this case the context of which mechanism of line relief is better suited to the realities of combat, to deduce the more likely translation.

In the case of Livy's legion, he explicitly limits ordines to the pilani, the troops under and behind the standards (triarii, rorarii and in this particular battle accensi).  The antepilani (hastati and principes) are organised in manipuli, not ordines.

We should incidentally note, gentlemen, that Livy's text (VIII.8.9) describing the relief of the hastati does not say 'inter ordines' but 'in intervalla ordinum', into the intervals of the files.

Latin "ordo" does not mean "row".

Actually it does.  Have a look at the 'ordo' entry in the Perseus lexicon, first meaning.

Present Italian usage is not a reliable guide, although it can occasionally confer insights.  Latin changed from the early Empire to the late Empire to the Vulgar Latin of Italy in the 10th century AD that served as the basis for the emergence of Italian in the 12th and subsequent centuries.  Dante's Commedia was the first widely propagated document in what we can recognise as the Italian language.  There are significant differences between Italian and Latin, not only in pronunciation (the Italian 'Ches-ah-re' for Caesar is much less accurate than the German 'Kaiser') but also in vocabulary (think of 'German' in Italian and Latin).  Arguing that a word has a certain meaning in Italian so must have meant the same in Latin is as unsafe as arguing that a word in Coptic infallibly demonstrates Twelfth Dynasty Egyptian (or even Ptolemaic) usage.

[andrew881runner]

I know but as an Italian I have always find translating from Latin pretty intuitive, because a lot of words kept the same root and same meaning. I am not saying that this happens for every Latin word, there are several Latin word Very different, but most are not. I have always noticed when translating that when I looked for all the meanings of the word I was confused, when I chosed the most similar Italian language I found the sense. This is most cases. So my choice of the most intuitive translation has its own meaning.
What you have said about history of Latin language and neo latin languages is very well known and totally true.

[Patrick Waterson]

I know but as an Italian I have always find translating from Latin pretty intuitive, because a lot of words kept the same root and same meaning. I am not saying that this happens for every Latin word, there are several Latin word Very different, but most are not. I have always noticed when translating that when I looked for all the meanings of the word I was confused, when I chosed the most similar Italian language I found the sense. This is most cases. So my choice of the most intuitive translation has its own meaning.
What you have said about history of Latin language and neo latin languages is very well known and totally true.

I expect you could be very helpful for giving us the 'feel' of intended meanings in Latin.  Do please feel free to do so when you find us discussing the meaning and usage of Latin words and expressions: you are the nearest thing we have to a native speaker on the subject. 

[Justin Swanton]

Looking at the Latin closely, I get a slightly different intepretation (sub iudicio!). First, for context, the preceding passage:

Clipeis antea Romani usi sunt; dein, postquam stipendiarii facti sunt, scuta pro clipeis fecere; et quod antea phalanges similes Macedonicis, hoc postea manipulatim structa acies coepit esse: postremo in plures ordines instruebantur. Ordo sexagenos milites, duos centuriones, vexillarium unum habebat. Romans formerly used the buckler, then, after they were put on pay, they replaced the buckler with the shield, and whereas they had formerly been Phalangites like the Macedonians, afterwards their lines were organised in a manipular fashion [manipulatim]. A company [ordo] had sixty soldiers, two centurions and one standard-bearer.

Notice that Livy defines what a company is. It is a group of 63 men, not a file.

Now for the passage in question. There is of course no punctuation in Latin, so one has to surmise from the context where the commas, semicolons and full stops go. This means ascertaining which subjects belong to which verbs and objects.

Prima acies hastati erant, manipuli quindecim, distantes inter se modicum spatium; manipulus levis vicenos milites, aliam turbam scutatorum habebat; leves autem, qui hastam tantum gaesaque gererent, vocabantur.

I cannot get 'The first line of battle were hastati, fifteen maniples, with a short distance between them and the maniple of twenty leves' out of the first two lines. There is no 'between' in the text. My interpretation is:

The first line were the Hastati, fifteen maniples with a small distance among/between them.

The 'small distance' must refer to the fifteen maniples. Question is: does it refer to gaps between one maniple and the next, or to gaps within the maniples? It could mean either. If the latter, these gaps would exist to receive the leves troops described below after they had done with skirmishing, before closing up in battle formation. It seems clear that Livy is using 'maniple' here as a synonym for 'ordo'.

Manipulus levis vicenos milites, aliam turbam scutatorum habebat. Leves autem, qui hastam tantum gaesaque gererent, vocabantur. A leves maniple (twenty men) had another group of shieldbearers. They were called 'Leves' who carried only a spear and javelins.

The Latin is rather obscure here (I suspect textual corruption). What Livy seems to be saying is that the 'Light' company had twenty light men and an additional group of heavier, shield-carrying troops.  The true 'light' troops carried only a spear and javelins, i.e. don't be fooled by the company's name.

Haec prima frons in acie florem iuvenum pubescentium ad militiam habebat. This frontmost part of the army had the freshest of the youth who had joined the military.

This refers to the twenty leves troops who would definitely need to hop, skip and jump in their role of skirmishers.

Robustior inde aetas totidem manipulorum, quibus principibus est nomen, hos sequebantur, scutati omnes, insignibus maxime armis. Those of a stronger age, called Principes, followed them in the same number of maniples. They all had shields and were the most heavily armed.

Hoc triginta manipulorum agmen antepilanos appellabant, quia sub signis iam alii quindecim ordines locabantur, ex quibus ordo unus quisque tres partes habebat. This force of thirty maniples were called 'antipilani' because under their banners another fifteen companies were placed. Of these each company had three parts.

Notice how Livy talks of 'another' fifteen ordines, in addition to the two groups of maniples already mentioned. This confirms he is using 'ordo' and 'manipulus' interchangeably.

Earum primam quamque primum pilum vocabant; tribus ex vexillis constabat; vexillum centum octoginta sex homines erant; primum vexillum triarios ducebat, veteranum militem spectatae virtutis, secundum rorarios, minus roboris aetate factisque, tertium [p. 460] accensos, minimae fiduciae manum; eo et in postremam aciem reiciebantur. The first of these parts was also called the primus pilus. One third [of the army] consisted of three banners. A banner had 186 men. The first banner led the Triarii, A force of older men of known ability, the second the Rorarii, of lesser strength in age and deed, the third the Accensi, the least capable, who were relegated to the last line.

This a different definition of the size of an 'ordo'. Each ordo is split into three banners (vexilla). Following the text literally, each banner has 186 men, which gives a total of 558 men for these rearward companies. Granted the the frontmost companies deploy 8 men wide by 8 deep, that will give a total depth of 88 men, which is wrong. Any theories?

Ubi his ordinibus exercitus instructus esset, hastati omnium primi pugnam inibant. si hastati profligare hostem non possent, pede presso eos retro cedentes in intervalla ordinum principes recipiebant. Tum principum pugna erat, hastati sequebantur. Where the army was formed up in these companies, the Hastati were the first to engage in battle. If the Hastati could not prevail against the enemy, they were received, as they fell back under pressure, into the gaps of the Principes' companies. The Hastati followed the Principes as they took up the fight.

Thus far Livy has defined an ordo as having 63 men (the size of a maniple) or 186 x 3 men (the size of nine maniples). Hence the 'gaps of the Principes' companies must, in context, refer to their maniples. The use of the genitive ordinum - suggests that these gaps were intrinsic to the companies, which is better understood as spaces between the files within a company rather than a gap between one company and the next. The Hastati then took their place behind the Principes and supported them - i.e. the engaged line was now 16 men deep.

Triarii sub vexillis considebant sinistro crure porrecto, scuta innixa umeris, hastas subrecta cuspide in terra fixas, haud secus quam vallo saepta inhorreret acies. The Triarii continued kneeling behind the ensigns, their left leg extended forward, holding their shields resting on their shoulders, and their spears fixed in the ground, with the points erect, so that their line bristled as if enclosed by a rampart.

Tenentes si apud principes quoque haud satis prospere esset pugnatum, gradun a prima acie ad triarios sensim referebant. inde rem ad triarios redisse, cum laboratur, proverbio increbruit. [12] triarii consurgentes, ubi in intervalla ordinum suorum principes et hastatos recepissent, [13] extemplo conpressis ordinibus velut claudebant vias unoque continenti agmine iam nulla spe post relicta in hostem incidebant. If the Principes also did not make sufficient impression in the fight, they retreated slowly from the front to the Triarii. Hence, when a difficulty is felt, "Matters have come to the Triarii," became a usual proverb. The Triarii rising up, after receiving the Principes and Hastati into the intervals of their companies, those companies then immediately pressed together as they, so to speak, closed the openings, in one compact body fell upon the enemy, no other hope being now left.

Notice how the Triarii immediately close up their companies after the Principes and Hastati have passed through them. This suggests not only file gaps (as opposed to spaces between maniples) but also wide file gaps - files standing 6 rather than 3 feet apart. The latter would not need to close up openings.

[Justin Swanton]

I would also direct attention to the specific meaning of 'inter ordinium' in Lucius Junius Moderatus Columella's Res Rustica III.13.3, IV.14.2 and V.5.3, which is 'the space between two rows'.  See the Perseus lexicon entry here.

True, but the expression is a little odd. Literally, it means 'between/among of the ordines', or perhaps more clearly: 'between/among that which is of the ordines.' 'That which is of the ordines' would correspond to rows, the component of the ordines and not the ordines themselves.

[Patrick Waterson]

I cannot get 'The first line of battle were hastati, fifteen maniples, with a short distance between them and the maniple of twenty leves' out of the first two lines. There is no 'between' in the text. My interpretation is:

The first line were the Hastati, fifteen maniples with a small distance among/between them.

The 'small distance' must refer to the fifteen maniples. Question is: does it refer to gaps between one maniple and the next, or to gaps within the maniples? It could mean either. If the latter, these gaps would exist to receive the leves troops described below after they had done with skirmishing, before closing up in battle formation. It seems clear that Livy is using 'maniple' here as a synonym for 'ordo'.

Manipulus levis vicenos milites, aliam turbam scutatorum habebat. Leves autem, qui hastam tantum gaesaque gererent, vocabantur. A leves maniple (twenty men) had another group of shieldbearers. They were called 'Leves' who carried only a spear and javelins.

The Latin is rather obscure here (I suspect textual corruption). What Livy seems to be saying is that the 'Light' company had twenty light men and an additional group of heavier, shield-carrying troops.  The true 'light' troops carried only a spear and javelins, i.e. don't be fooled by the company's name.

I see it as more 'Livian' than obscure.  What he is telling us, or trying to (assuming he follows his source correctly, which is debatable) is that each hastati maniple consists of 20 leves and the balance of 'scutati', the pilum-wielding, scutum-carrying troop type we traditionally regard as hastati.

That these maniples must be hastati becomes clear when we do the sums for legion strength as a whole.  The rear-line 'ordines' provide 15x180 (or 186) = 2,700 (or 2,790) legionaries out of 5,000-ish.  This leaves 2,300-ish to come form the first 30 maniples, half of whom are principes.  Dividing 2,300 by 30 gives 76.67 men per maniple, suggesting 80 men per maniple (as opposed to 60 per vexillum in the 180-strong ordo).  We can thus see that a maniple (of 80 men) is not interchangeable with a ordo (of 60 men), and it may be instructive that Livy never refers to an 'ordo' of hastati or principes in this description (they are always 'manipuli') but always uses 'ordo' to denote the pilani, the assemblage of triarii, rorarii and accensi under the standards.

Notice how Livy talks of 'another' fifteen ordines, in addition to the two groups of maniples already mentioned. This confirms he is using 'ordo' and 'manipulus' interchangeably.

Or perhaps not, as explained above.  I would not read too much into that single 'alii'.

Earum primam quamque primum pilum vocabant; tribus ex vexillis constabat; vexillum centum octoginta sex homines erant; primum vexillum triarios ducebat, veteranum militem spectatae virtutis, secundum rorarios, minus roboris aetate factisque, tertium [p. 460] accensos, minimae fiduciae manum; eo et in postremam aciem reiciebantur. The first of these parts was also called the primus pilus. One third [of the army] consisted of three banners. A banner had 186 men. The first banner led the Triarii, A force of older men of known ability, the second the Rorarii, of lesser strength in age and deed, the third the Accensi, the least capable, who were relegated to the last line.

This a different definition of the size of an 'ordo'. Each ordo is split into three banners (vexilla). Following the text literally, each banner has 186 men, which gives a total of 558 men for these rearward companies. Granted the the frontmost companies deploy 8 men wide by 8 deep, that will give a total depth of 88 men, which is wrong. Any theories?

Actually the Livian maniple (manipulus) would deploy 10 wide and 8 deep, a natural configuration for 80 men (and one which allows the 20 leves to back-rank the 60 hastati scutati once skirmishing is done).

Livy lists an ordo as 186 men because he is adding in two officers per vexillum.  He lists a vexillum as 60 men because that is how many men it had.  One suspects he is following his own source statements without seeing the apparent inconsistency.  Triarii (and similarly accensi, although I suspect they would normally just back up triarii) would fight 10 wide and 6 deep; rorarii, when committed, probably the same.

Depth is thus 8 men for each maniple line (hastati and principes) and 18 for the ordo line (triarii, rorarii and accensi).  The total depth is thus 16+18 = 34 men.

Ubi his ordinibus exercitus instructus esset, hastati omnium primi pugnam inibant. si hastati profligare hostem non possent, pede presso eos retro cedentes in intervalla ordinum principes recipiebant. Tum principum pugna erat, hastati sequebantur. Where the army was formed up in these companies, the Hastati were the first to engage in battle. If the Hastati could not prevail against the enemy, they were received, as they fell back under pressure, into the gaps of the Principes' companies. The Hastati followed the Principes as they took up the fight.

Thus far Livy has defined an ordo as having 63 men (the size of a maniple) or 186 x 3 men (the size of nine maniples). Hence the 'gaps of the Principes' companies must, in context, refer to their maniples. The use of the genitive ordinum - suggests that these gaps were intrinsic to the companies, which is better understood as spaces between the files within a company rather than a gap between one company and the next. The Hastati then took their place behind the Principes and supported them - i.e. the engaged line was now 16 men deep.

Indeed.  Though as Livy never uses 'ordo' to represent a maniple, as demonstrated above, I rather suspect his 'in intervalla ordinum' refers directly to the gaps between files.  The essential conclusion is the same, anyway. 

Tenentes si apud principes quoque haud satis prospere esset pugnatum, gradun a prima acie ad triarios sensim referebant. inde rem ad triarios redisse, cum laboratur, proverbio increbruit. [12] triarii consurgentes, ubi in intervalla ordinum suorum principes et hastatos recepissent, [13] extemplo conpressis ordinibus velut claudebant vias unoque continenti agmine iam nulla spe post relicta in hostem incidebant. If the Principes also did not make sufficient impression in the fight, they retreated slowly from the front to the Triarii. Hence, when a difficulty is felt, "Matters have come to the Triarii," became a usual proverb. The Triarii rising up, after receiving the Principes and Hastati into the intervals of their companies, those companies then immediately pressed together as they, so to speak, closed the openings, in one compact body fell upon the enemy, no other hope being now left.

Notice how the Triarii immediately close up their companies after the Principes and Hastati have passed through them. This suggests not only file gaps (as opposed to spaces between maniples) but also wide file gaps - files standing 6 rather than 3 feet apart. The latter would not need to close up openings.

True, especially as the rorarii would earlier have had to move through the triarii in order to make their contribution.  I think 6' gaps for triarii files is implied both by this passage and by battlefield convenience.

Conversely, the principes might not need to have 6' gaps to let the hastati through; 3' intervals would be sufficient and would have the advantage of allowing mid-combat hand over without the need to race alternate files into position while one's front rankers were being double-teamed by the opposition.

[Erpingham]

Two thoughts have struck me considering the line relief problem again.

Firstly, in wargames terms, the scale of the representation may mean the exact mechanism doesn't matter.  If working at a high enough scale, an abstraction such as allowing the principes to interpenetrate a unit of principes who are close enough behind them may be enough.  Similarly between the antepilani and the triarii.  If fighting at a more detailed level, would line relief be carries out at a level lower than a legion?  Indeed, if fighting at a level of less than a legion, would all three lines be there at all?

Second thought is more a question.  Our detailed examples of line relief are quite early in the evolution of the army.  At what point does line relief stop?  Or does it not stop but evolve.  Caesar's legions still seem to have three lines but the third line appears to be a tactical reserve rather than a final line of defence.  Lines one and two do interconnect.  Indeed, Roy has suggested in another thread that lines one and two at this point are one tactical entity which enter the fray together.  So do we have line relief at this point?  In the Empire, multiline deployments continue but does any system of line relief?

[Justin Swanton]

That these maniples must be hastati becomes clear when we do the sums for legion strength as a whole.  The rear-line 'ordines' provide 15x180 (or 186) = 2,700 (or 2,790) legionaries out of 5,000-ish.  This leaves 2,300-ish to come form the first 30 maniples, half of whom are principes.  Dividing 2,300 by 30 gives 76.67 men per maniple, suggesting 80 men per maniple (as opposed to 60 per vexillum in the 180-strong ordo).  We can thus see that a maniple (of 80 men) is not interchangeable with a ordo (of 60 men), and it may be instructive that Livy never refers to an 'ordo' of hastati or principes in this description (they are always 'manipuli') but always uses 'ordo' to denote the pilani, the assemblage of triarii, rorarii and accensi under the standards.

Indirectly he does. The 'other fifteen companies' (ordines) each consist of one vexillum of Triarii, one of Rorarii and one of Accensi. He does not call each one a company, true, the reason being that they were considered to form a single entity, the 'one third' of the army composed of banners (vexilla), and not separate lines like the Hastati and Principes.

Livy lists an ordo as 186 men because he is adding in two officers per vexillum.  He lists a vexillum as 60 men because that is how many men it had.  One suspects he is following his own source statements without seeing the apparent inconsistency.  Triarii (and similarly accensi, although I suspect they would normally just back up triarii) would fight 10 wide and 6 deep; rorarii, when committed, probably the same.

My problem was that each vexillum seems to have 186 men, which is far too many. But taking a closer look at the Latin: vexillum centum octoginta sex homines erant. The 'erant' (plural) has 'homines' as subject, not 'vexillum' (singular). Using a plural verb in this manner with the object in the beginning of the sentence seems a bit strange to my (admittedly inexperienced) eye. Correct vexillum to vexilla and the Latin flows much more smoothly: "The banners were 186 men', i.e. the three banners of the rear company totalled 186 men, each banner 62 men, and everything then clicks into place. Just a suggestion.

I think 6' gaps for triarii files is implied both by this passage and by battlefield convenience.

Conversely, the principes might not need to have 6' gaps to let the hastati through; 3' intervals would be sufficient and would have the advantage of allowing mid-combat hand over without the need to race alternate files into position while one's front rankers were being double-teamed by the opposition.

If the Triarii are arranged with wide gaps between the files it makes sense that the Principes and Hastati are similarly arranged, since the same relief process happens with them (the Hastati receiving the skirmishers). Seems more logical. With my proposed relief mechanism the front rankers would not face more than one opponent at any time.

[Patrick Waterson]

Two thoughts have struck me considering the line relief problem again.

Firstly, in wargames terms, the scale of the representation may mean the exact mechanism doesn't matter.  If working at a high enough scale, an abstraction such as allowing the principes to interpenetrate a unit of principes who are close enough behind them may be enough.  Similarly between the antepilani and the triarii. 

At that level of abstraction, a simple exchange of one line of figures for another would indeed suffice.  The precise mechanism is primarily of historical interest so we can - hopefully - understand how this particular army worked.  This in turn helps to depict it and its functioning with greater accuracy when attempting less abstract systems - and also to learn why it fared as it did against particular opponents.

If fighting at a more detailed level, would line relief be carries out at a level lower than a legion?  Indeed, if fighting at a level of less than a legion, would all three lines be there at all?

Actually line relief seems to have been an army-level thing not a legion-level thing.  There are few instances of a single legion fighting on its own and as far as I know none of these detail what it did or how.  Until the Empire a legion tended to be regarded as indivisible, although in the 1st century BC there was a tendency for a commander to have so many 'cohorts' which might or might not be partially or fully organised into legions.  Cohorts seem to have fought as discrete entities rather than splitting into three lines.

Second thought is more a question.  Our detailed examples of line relief are quite early in the evolution of the army.  At what point does line relief stop?  Or does it not stop but evolve.  Caesar's legions still seem to have three lines but the third line appears to be a tactical reserve rather than a final line of defence.  Lines one and two do interconnect.  Indeed, Roy has suggested in another thread that lines one and two at this point are one tactical entity which enter the fray together.  So do we have line relief at this point?  In the Empire, multiline deployments continue but does any system of line relief?

Answering this one will require a bit of a trawl through our sources from Tactius to Zosimus; in the 4th century AD, which is Ammianus' era, the legions have already been diminished (perhaps by Constantine) to about one third of their former size (and multiplied in number accordingly).  In the one battle Ammianus describes in detail (Argentoratum - he says a lot about Adrianople but rather less to the point) there is commitment of reserves (the Cornuti and Bracchiati) but no apparent line relief as such.  Whether the reserves were able to conduct a traditional line relief and substitute for failing units in the front line is an open question.

Tacitus, writing about AD 69, gives us a lively description of the Second Battle of Bedriacum, which was distinguished by one side having leadership and the other none (having just put their commander in chains on the not unjustified suspicion of treason), and the unit of action seems to be the legion, not part of a legion.  Whether this terminology conceals or assumes internal line relief is an unanswered question.

Writing in the 4th century AD, Vegetius observes (III.16, Reserves):

"The method of having bodies of reserves in rear of the army, composed of choice infantry and cavalry, commanded by the supernumerary lieutenant generals, counts and tribunes, is very judicious and of great consequence towards the gaining of a battle. Some should be posted in rear of the wings and some near the centre, to be ready to fly immediately to the assistance of any part of the line which is hard pressed, to prevent its being pierced, to supply the vacancies made therein during the action and thereby to keep up the courage of their fellow soldiers and check the impetuosity of the enemy. This was an invention of the Lacedaemonians, in which they were imitated by the Carthaginians. The Romans have since observed it, and indeed no better disposition can be found."

The absence of line relief is noticeable, and would seem to be the norm for the 4th century AD.  However in II.10 (Drawing up a Legion in Order of Battle) Vegetius describes the 'ancient legion', which appears to be an updated and rearmed version of the late Republican legion, and may in fact be a description of the pre-4th century Imperial legion (much ink has been spilt on this point).

"We shall exemplify the manner of drawing up an army in order of battle in the instance of one legion, which may serve for any number. The cavalry are posted on the wings. The infantry begin to form on a line with the :first cohort on the right. The second cohort draws up on the left of the first; the third occupies the center; the fourth is posted next; and the fifth closes the left flank. The ordinarii, the other officers and the soldiers of the first line, ranged before and round the ensigns, were called the principes. They were all heavy armed troops and had helmets, cuirasses, greaves, and shields. Their offensive weapons were large swords, called spathae, and smaller ones called semispathae together with five loaded javelins in the concavity of the shield, which they threw at the first charge. They had likewise two other javelins, the largest of which was composed of a staff five feet and a half long and a triangular head of iron nine inches long. This was formerly called the pilum, but now it is known by the name of spiculum. The soldiers were particularly exercised in the use of this weapon, because when thrown with force and skill it often penetrated the shields of the foot and the cuirasses of the horse. The other javelin was of smaller size; its triangular point was only five inches long and the staff three feet and one half. It was anciently called verriculum but now verutum.

The first line, as I said before, was composed of the principes; the hastati formed the second and were armed in the same manner. In the second line the sixth cohort was posted on the right flank, with the seventh on its left; the eighth drew up in the center; the ninth was the next; and the tenth always closed the left flank. In the rear of these two lines were the ferentarii, light infantry and the troops armed with shields, loaded javelins, swords and common missile weapons, much in the same manner as our modern soldiers. This was also the post of the archers who had helmets, cuirasses, swords, bows and arrows; of the slingers who threw stones with the common sling or with the fustibalus; and of the tragularii who annoyed the enemy with arrows from the manubalistae or arcubalistae.

In the rear of all the lines, the triarii, completely armed, were drawn up. They had shields, cuirasses, helmets, greaves, swords, daggers, loaded javelins, and two of the common missile weapons. They rested during the action on one knee, so that if the first lines were obliged to give way, they might be fresh when brought up to the charge, and thereby retrieve what was lost and recover the victory. All the ensigns though, of the infantry, wore cuirasses of a smaller sort and covered their helmets with the shaggy skins of beasts to make themselves appear more terrible to the enemy. But the centurions had complete cuirasses, shields, and helmets of iron, the crest of which, placed transversely thereon, were ornamented with silver that they might be more easily distinguished by their respective soldiers."

On balance, this would seem to better describe an Imperial legion despite the Republican (or presumed Republican) terminology.  If so, then line relief is implicit in the descriptions, though not explicitly described in the brief paragraph that follows:

"The following disposition deserves the greatest attention. In the beginning of an engagement, the first and second lines remained immovable on their ground, and the trairii in their usual positions. The light-armed troops, composed as above mentioned, advanced in the front of the line, and attacked the enemy. If they could make them give way, they pursued them; but if they were repulsed by superior bravery or numbers, they retired behind their own heavy armed infantry, which appeared like a wall of iron and renewed the action, at first with their missile weapons, then sword in hand. If they broke the enemy they never pursued them, least they should break their ranks or throw the line into confusion, and lest the enemy, taking advantage of their disorder, should return to the attack and destroy them without difficulty. The pursuit therefore was entirely left to the light-armed troops and the cavalry. By these precautions and dispositions the legion was victorious without danger, or if the contrary happened, was preserved without any considerable loss, for as it is not calculated for pursuit, it is likewise not easily thrown into disorder."

In essence, we can say that line relief probably continued under the Empire, but our sources lack sufficient detail to be sure.

[Erpingham]

Thanks Patrick.  I personally find Vegetius difficult in his description- is he describing anything that really existed or is he collating a variety of sources into what looks like a whole?  He clearly has a description of an early Republican legion (possibly the same one as we have) and he knows how modern legionaries are equipped but does he have a real lost manual on the operation of the army in between?  (I think to answer that we would need a major new theme )

I don't think Arrian's formation against the Alans features three line legions and line relief, but I could be wrong.  But that would give us an example between Bedriacum and Argentoratum.

Anyway, plenty of scope I think in the evolution between the earlier and later Republican (which we do have info on) to see how things were changing and what trends there might be.

[Jim Webster]

One thing that I've not noticed anybody comment on is 'Who ordered line relief to take place?'

Was it ordered at the level of the Legion or the level of the Maniple, because there isn't really any level of command in between, unless it was a job assigned to tribunes

Jim

[Justin Swanton]

One thing that I've not noticed anybody comment on is 'Who ordered line relief to take place?'

Was it ordered at the level of the Legion or the level of the Maniple, because there isn't really any level of command in between, unless it was a job assigned to tribunes

Jim

It wasn't necessarily ordered. Assume the Hastati are being given a hard time and yield ground to their opponents. After a while the rearmost rank of the Hastati contacts the front rank of the Principes. At this point the Hastati files automatically start streaming back through the Principes lanes (it is something they have trained for and execute when the occasion arises). The same thing happens for the Principes (now backed by the Hastati - they file through the Triarii the moment they make contact with them, being driven back by the enemy.

[aligern]

I am afraid that is a misreading of Roman history Justin. These things are done on command and only on command . The Romans go to a lot of trouble to give examples if how even the most socially superior must submit to discipline  and command.

A general just cannot have the men doing their own thing, it leads to disaster.  The hastati have to hold until they are told to fall back, otherwise it will happen at different tomes in different areas and the  whole legion will be put in danger. In the example that we have from Caesar the whole third line of the army is ordered up...that is an army level order.

Roy