walter woodburn hyde - 1924 - traians danube road and bridge

7/30/2019 Walter Woodburn Hyde - 1924 - Traians Danube Road and Bridge

1/7

Trajan's Danube Road and Bridge

Author(s): Walter Woodburn HydeSource: The Classical Weekly, Vol. 18, No. 8 (Dec. 8, 1924), pp. 59-64Published by: Classical Association of the Atlantic StatesStable URL: http://www.jstor.org/stable/4388602Accessed: 05/11/2009 17:47

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=classaas.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Classical Association of the Atlantic States is collaborating with JSTOR to digitize, preserve and extend access

to The Classical Weekly.

http://www.jstor.org
http://www.jstor.org/stable/4388602?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=classaashttp://www.jstor.org/action/showPublisher?publisherCode=classaashttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/4388602?origin=JSTOR-pdf


2/7

DECEMBER 8, 1924] THE CLASSICAL WEEKLY 59TRAJAN'SDANUBE ROAD AND BRIDGE

I was greatly interested in the excellent study by M\r.Benjamin W. Wells of Trade and Travel in the RomanEmpire I, which appeared in The Classical Journal19.7-i6 (October, 1923). In one passage the authormentions the rock-cut road built by Trajan along theSerbian bank of the Danube, above the Iron Gate, as amilitary approach to Dacia. In speaking of how theimperial engineers solved apparently insurmountabledifficulties, he says (i i): ". . .A road was carriedthrough the Iron Gates as far as Orsova, at times onbeams fastened to the sheer face of the cliffs and over-hanging the river". However, the rapids known as theIron Gate begin five miles below Orsova, while therock part of the road in question is above that town.Many historians of Trajan's Dacian Wars have similar-ly mistaken the localities involved. Thus Duruy, in hisHistory of Rome (V, Pt. I, 234), regards the Iron Gateas comprehending that part of the Danube-includingthe defile of the Kazan-which extends from Drenkovato near Orsova, a distance of over fifty miles. MIerivale(History of the Romans under the Empire, 7.I90)makes the same mistake when, in speaking of Trajan'sroad, he says it "threaded the defile of the Iron Gate".For, while rightly placing the Iron Gate below Orsova,he wrongly locates the road in question. As I havevisited the locality twice, perhaps a brief account of theroad above the Iron Gate, and of Trajan's bridge belowit, and of the use made of both in the Dacian Wars maybe of service. In September, I898, while en route forConstantinople, I started down the river by steamerfrom Belgrade, but, because of unprecedented lowwater, was compelled to disembark at the village ofOmoldova, six and a half hours below Belgrade, andcontinue the rest of the way to Orsova, some fiftymiles beyond, on a peasant's springless cart along theZ6chenyi road on the Hungarian bank of the river.Again, in August, I907, I reached Orsova from Bucha-rest, and took the steamer against the current to Bel-grade. On both occasions, and especially on the latter,I had excellent opportunities to view the rock-hewnroad at close range.

Trajan's desire to restore the splendor of Roman armsseems to have been largely responsible for his idea ofstrengthening the chain of frontier defences on theRhine and the Danube. Thus, if we ate to believeAurelius Victor', the Roman historian of the Caesarswho lived in the reigns of Constantius and his successors,Trajan constructed a chain of military roads from theMain to the Black Sea-iter conditum per ferasgentes, quo facile ab usque Pontico mari in Galliampermeatur. But the building of so vast a series ofworks must have been the labor of many years,beginning long before the Emperor's time. As gover-nor, Trajan had already spent three years on thedefences of the Rhine and the agri decumates, andlater, after his accession, recognizing that the center ofgravity of the Empire was further east, he turned hiisattention to the Danube. Here and there traces ofthese undertakings are visible. Thus, portions of awall ascribed to Traian still exist above Regensbur,.

extending from Eining, near the anicient Abusina, toWiesbaden, while another linie of fortifications can betraced on the lower Danube in the Dobrudja betweenCernovada and Constantza. At intervening pointsmany other Roman remains have been found, notablythe scar left by the rock-hewn road in the defile of theKazan, and the foundations of the hridge below theIron Gate at Turnu Severinu. But the long course ofthie ancient line of defence to-day is far better markedby the surviving names of military posts and coloniesalong it.

The remnants of the road in question are first prom-inent below the Serbian village of Golubacs-theancient Cuppae-,which lies almost due south ofOmoldova across the great island formed in thisneighborhood by the forking of the Danube. Here isthe picturesque ruin of a castle high on a rock, itsseven towers visible for a long distance, the old fortressof Columba, or the 'Dove', which has given its name tothe modern hamlet. It was here that the GreekEmpress Helena was once incarcerated, and thisfortress has been the scene of many a struggle betweenChristian and Turk. From Belgrade to this point theDanube is open and wide, in places reaching a width oftwo miles. its low-lying banks covered with forests andmorasses. But the scenery gradually changes, as theriver enters the highland district where the southern-most spurs of the Carpathians-the Szretinye moun-tains-rea ch the river in the great promontory ofK-azan, and confront the northernmost outliers of theBalkans. For miles the river, which in geologic agesburst through these foothills, is confined by loftycliffs in a narrow channel. Here begins the finest part ofthe Danube scenery, which is wild and impressive, therocky banks at times being precipitous, and againsloping or broken by torrents. AMoreover,a series ofrapids, whose swirling waters beat against the cliffwalls and remind us on a smaller scale of the rapids inthe Niagara River, has, until recently, impeded navi-gation. The Hungarian government, as the resultultimately of the action of the Great Powers at theCongress of Berlin, in I878, carried out at great expenseengineering works intended to make the whole stretchof the river from Omoldova to Turnu Severinu, a dis-tance of ovei sixty miles, navigable in all seasons, bymaking the bed of the river at low water at least twometers deep above Orsova, and three below it.

The first rapids of the series, known as the Stenka andcontinuing for i,ioo yards, are just above Drenkova, atown some miles below Omoldova. Here a rock bank,partly exposed, extended nearly across the river, whichat this point is 985 yards in width. Nine miles beloware the Kozla Dojke rapids of similar length, where twobanks of rock caused a sudden alternation in the current,the river narrowing to a width of I 70 to 330 yards.Two more rapids follow in close succession, the Isla'zand Tachtalia, and six miles below Kozla Dojke arethose of the Greben, which formed the most seriousimpediment of all to navigation. For here a spur of theGreben hills projected between two shoals, and theriver narrowed to 300 yards, only to widen almostimmediately to one and one-half miles. In the GrebenDe Caesaribus 13.


3/7

60 THE CLASSICAL WEEKLY [VOL. XVIII, No. 8, WHOLE No. 485rapids are the remains of what appears to have been aRoman canal cut into the river bed. Seven milesfurther down are the Jucz rapids. The Austro-Hunga-rian engineers changed all these adverse conditions toriver traffic by cutting canals 66 yards wide and six andone-half feet deep into the rock bottom of the riverthrough the first four rapids, by cutting back for I67yards the original face of the Greben rock and buildingbelow it a wall four miles in length along the Serbianshore, and by constructing both a canal and a buttresswall in the Jucz.

Below the last named rapids the river again widens tonearly a mile, and we soon pass the Serbian villages ofMilanovafcs and Tricule-the latter marked by theruins of three Roman towers. It is in this neighbor-hood that the famous pass known as the Klisura orKazan-the latter word meaning 'kettle'-begins,some twenty miles above Orsova. The river's course isovershadowed by high cliffs, which at one point on theright bank form a mighty bastion over 2,000 feet inheight, known as the Sterbeczu Almare. The narrowestpart of the defile is opposite the Hungarian village ofRogach, where the river contracts to only I62 yards inwidth, and has a tremendous current, and is even saidto reach the incredible depth of I50 feet.

It is here in the Kazan that the best traces of Tra-jan's road are seen cut into the face of the perpendicularcliffs. These remains are confined, for the most part, toa horizontal ledge which is from four to six feet inwidth and considerably higher, hewn from the rock afew feet above the river. However, a wvidthat leasttwice as great was gained for the roadway by a woodengallery built out over the water and supported by aframework of beams. Below the cutting are plainlyvisible the holes in the rock, placed at irregular inter-vals, into which the horizontal and slant beams werefastened which upheld the wooden platform. Theseholes are about nine inches square and eighteen indepth, and are placed at four distinct elevations.Chasms in the rock-wall were spanned by bridges,either of masonry or wood, of which there are notraces. It is not difficult to understand why only aportion of the road was hewn in the rock. The woodengallery was not constructed merely to save expense,btit rather to furnish a convenient tow-path for thepassage of transports, and, furthermore, it couldreadily be destroyed in times of danger.

The defile of the Kazan ends opposite the squalidvillage of Ogradina on the north bank, some distancesouthwest of Orsova. Here, on the Serbian side, is theTabula Trajana, as the modern inscription on theoverhanging rock above it is called, one of the most im-portant historical monuments along tlhe Danube.The letters are cut into the limestone, the tablet beingpelhaps twenty feet above the water. While the in-scription has been somewhat obscured by weatheringand especially by the smoke from the fires of theSerbian fisherman, who for a long time have used thespot for their camps, it can still be fairly well made out.In recent years the level of the roadway has beenraised by an inclined plane of masonry as a protectionagainst the use made of it by boatmen. I quote a

description of this tablet, which has been executedwith some pretence to artistic elegance, from JohnPaget, who visited the region in the thirties of the lastcentury2:"A winged genius on each side supports anoblongtablet protectedby the overhanging ock,whichhas been carved into a rock cornice, surmountedbya Roman eagle. At either end is a dolphin". Theinscriptioncontains only six lines and reads:IMP . CAESAR . DIVI . NERVAE . FNERVA TRAIANUS . AVG. GERMPONTIF MAXIMVS TRIB POT IIIIPATER PATRIAE COS IIIMONTIBUS EXCISIS AMNIB VSSVPERATIS VIAMI FECIT3It mentionsTrajanas the builderof the road, and alsoindicates the approximatedate of its completion. Forin it Trajan is called Germanicus,a title which, weknow, he and Nerva received in 97 A. D., but notDacicus, a title which he received in the fall of I02,some months after the end of the First Dacian War.Furthermore, he Emperor s calledconsulfor the thirdtime, and tribunefor the fourth. We know that he re-ceived his first consulshipin 9I, the second in 97, andthe third in 99, the year after he became Emperor,and that he received his first tribunate in 97, and hisfourthin Ioo. Consequentlythe tablet, in honorof thecompletionof the road, must have been erectedsome-time in IOOA. D., and in any casebeforethe end of I02.Hence we may reasonablyconcludethat the road mayhave been ready for use or nearly so before the out-break of the First Dacian War in the spring of ioi.As the great bridge further down the river, to whichthe road later was extended,seems to have been begunin I03, Trajanmay have had in mind the plan of con-structing both road and bridge before engagingin hisfirststrugglewiththe Dacians.Until the Szechenyi road was completed in i885from the town of Bazias, which is an hour and a halfabove Omoldova on the north bank, to Orsova, theformer rontiertown of Hungary,the Kazan,we may besure, had not been traversed by vehicles since soonafter Trajan'sday. And beforethat time we knowthatnoroadhadbeencompleted hroughthe defile,althoughan attempthad beenmadeto constructone. Thus, justbelow the rapids of Kozla Dojke, one still sees a plaintablet with mutilated lettering cut into the rock on theSerbianside, which Mommsendated in the year 33 or34 A. D.4 It records that two Roman legions,thefourth Scythian and the fifth Macedonian,had beenemployedby the Emperor Tiberius in cutting a roadat this point, therefore t least 67 years before Trajanfinally carried the project to completion.At Orsova, some distance beyond the exit of thedefile,the surfaceof the river is 4.2 feet above the sea.From here the channel widens once more to nearly amile, the river now having abruptly changedits direc-

'Hungary and Transylvania (London, i839, Philadelphia, I850.See Volume i I, page 37 of the Philadelphia edition).'C. I. L. III, I, No. I699. Mommsen reads the last two lines asthey are given above. The first publisher of the inscription, inMittheilungen der Zentralcommission zur Erforschung undArneth,Erhaltung von Baudenkmaeler, 83 f. (Vienna, I856), read them asMONTIS E FLUVII ANFRACTIBVSSVPERATIS VIAM PATEFECIT.4C. I. L. III, I, No. i698.


4/7

DECEMBER 8, 1924] THE CLASSICALWEEKLY 61tion, which from Milanovacs had been northeast, tosoutheast, till five miles beyond, or 27 below the Juczrapids, the last rapids in its course begin, the so calledIron Gate. The Turkswere responsible or this name,which is the translation of Demir Kapu. They gave itto this locality not because of high banks along theriver, sincethese have graduallyapproachedo near thelevel of the water, but because of the fact that untilrecently the Prigrada rock bank nearly blocked theriver'scourseat this point. In consequenceof the name,these rapids have long had a sinister reputation as adangerousbar to navigation, even though they havealways been shot by ordinary open fishing-boats.The chief obstacle was rather the shallowness of thewater. This difficultyhas now been obviated by thecutting of a canal in the bed of the river near theSerbian shore, ten feet deep, eighty yards wide, andsome 2,000 yards in length. In this neighborhood heremains of a smaller Roman cutting in the rock bedwere long known. Buttress walls have also been con-structedon either bank to restrainthe swirlingwaters.This workwas completedon OctoberI, I898, just a fewdays beforeI took my first Danube journey. In conse-quence of these improvements the Iron Gate is nowalways open to navigation,except sometimesin winterwhen it is chokedby ice. At the exit of the Iron Gate-whichis nearly two miles in length-the river flows onsmoothly, its low banks now separatingnortheasternSerbia romRumania. It wasin this neighborhoodhatTrajanbuilt his bridge.

Of all Trajan'smilitary bridges, including hose overthe Rhine near AMayence,t Chemtou in Tunis, andtwo in Spain, one of which crossed the Tagus atAlcantara,and that over the Tamango at Chaves inGalicia,this one across the Danubewas by far the mostimpressive. We have a very brief account of it inXiphilinus's epitome of Book 68 of Cassius Dio'sRoman History5. As the Ronman istorian was legateof part of Pannonia in the year 227 A. D., he mayactuallyhave seen the bridge in its then state of ruin.He says it was supportedby twenty stone piers,whichwere I50 feet high,6o broad,and I70 apart; the pierswere connected by arches. He adds that in his dayonly the piers remained standing, and afforded nomeans of crossing. Therefore t seemedto him 'as ifthey were erectedfor the sole purposeof demonstratingthat there is nothing which human energy cannot ac-complish'. Both Dio and the Byzantine historianProcopius record that Apollodorusof Damascus wasthe architect, the latter also adding that in his day,in the sixth century, the channel of the river was ob-structed by the ruins6. Dio says further that Apollo-dorus planned Trajan's Forum with its buildings atRome, and both he and Spartian recordhis death at thehandsof Hadrian7. AureliusVictor,who was governorof part of Pannonia under Julian, and Tzetzes, thetwelfth century grammarian of Constantinople, giveother detailsabout thebridge, he latter mentioning hat

Apollodorus constructedan island in the river's bed8.If the architect's account of his work had survived,wecould more easily refute some of Dio's exaggerations.rnstead of the 'violent and deep' current of the river,whose channel is 'full of eddies', the Danube at thispoint is nearly three-quartersof a mile wide and rela-tively shallow, and its current is fairly gentle. Ofcourse conditions may have somewhat changed sinceDio's time, sinceit is well knownthat the Danube is anexample of what the veoloristscall a dying river. Theheight of the bridge, I45'2 English feet9, seems alsoexaggerated, since the low banks on either side wouldhave made such an elevation not only unnecessary,but objectionable. To be sure, he bridge at Alcantara,which is still standing, is 197 feet high, though only6i6 feet long with its six arches'0,but there such aheight was absolutely necessary. Only in the numberof piers, and in their width and intervals, can Dio beshown to have been nearer the truth. In any case,Trajan'sDanube bridgewas a unique feat of ancientconstruction,and has not been duplicated n this partof the world n moderndays, since the last bridge downthe Danube is still the one built between Buda andPesth in westernHungary, some 300 miles upstream.The remains of the bridge can still be seen in the bedof the river at low waterjust above TurnuSeverinu inRumania, a town near the site of the Roman outpostDrobetae, the first station on the road which Trajanlater constructed between the Danube and Apulum(Karlsburg)in Upper Dacia. Turnu Severinu is be-tween six and seven miles below the exit of the IronGate, and consequentlyabout thirteen below Orsova.Here,around he year 246A. D., Severinus,governorofMoesiaunder Philip the Arab, built the TurrisSeverina,which gave to little Wallachiadown to I247 its nameTerra de Severinu. Near this point, then, the hugebridge was thrown over the river to Egeta on theopposite bank, where, down to the fifth century, atleast a part of the thirteenth legion (Gemina)and ofthe river fleet was stationed. At present the Turkishvillage of Feth Islam-very near the larger town ofKladovo-occupies the site of Egeta. All that is nowleft of the bridgeis a shapelessmass of masonrysometwenty feet high on either bank and betweenthese andthe water's edgebroken wallsrisingto the height of thebanks of the river. These structures are doubtlesstheremains of the buttressesfrom which the end archesofthe bridge sprang. And in the bed of the river in a linebetween these ruins areclearlyvisible,projecting wo orthree feet abovethe surfacein low water,the remnantsOf he foundationpiers. Near the middle of the river isa low island which occupies the space of perhaps ourPiers; some believe this to be the remnant of the arti-ficial island constructed by Apollodorus. Froehnershows nineteen such substructuresexclusive of the endbuttresses"1. Even fragments of the wooden super-

568.I3.I-6. Wiphilinus wrote in the reign of Michael VII (Ducas),I07I-I078 A. D.5De Aedificiis 4. 6.I6.7Dio 69.4: Spartian, Vita Hadriani i9.

BVictor,De Caesaribus I3.4; Tzetzes, Chiliades 2.67.9We gain this figure by using Ddrpfeld's computation of theGraeco-Roman foot as .296 of a meter, or about II.65 Englishinches: see Athenische Mitteilungen, 7 (1882), 282."?SeeDuruy, History of Rome, V, Pt. r, cut opposite page 274;and C. I. L. II, Nos. 759 and 762."See diagram in La Colonne Trajane', (i865), opposite pageI33 (the second edition, with one volume of text and four of plates,


5/7

62 THE CLASSICAL WEEKLY [VOL. XVIII, No 8, WHOLENo. 485.tructure were found in I858 by Aschbach, who was incharge of the Austrian engineers who carefullyTstudiedthe remains12. The intervals between the piers, whichDio says were 170 feet-or I651 / English feet-probably reckoned between the axes, average 17014feet, or from the outer corners II0o' feet. The pierswhich Dio gives as sixty feet-581,' English feet-Inwidth are really, accordinp to Froehner"3, from 45-47feet long and from 69-72 feet wide. Such enormouspiles of masonry, constructed of op?s incertunm.and en-cased in regular courses of hewn stone, were deemednecessary not only to uphold the arched roadway, butto witlstand the pressure of water and ice.

Besides these meager remains we also have severalrepresentations of the bridge on Roman works of art,which show that the framework of arches and roadwaywas of wood, as indeed we should expect for spans ofsuch size. Thus the bridge is partly shown on one ofthe scenes from the band of reliefs wvhich till decoratesthe Column of Trajan at Rome. Here is depicted asacrifice which is apparently in commemoration of thecompletion of the structure. In the foreground standthe Emperor and several others-among wvhomCicho-rius believes he can identify Hadrian, wvhoat that timewas one of Trajan's generals, and Apollodorus, thearchitect-around an altar, where a priest standsready to slay a sacrificial ox. In the background wesee five square pillars and five arches of the bridge, andthe roadway with its long balustrade of wood14. Thisroadway is reached from the Moesian side by astairway leading up through a monumental gateway,while at either end of the structure a fortress is repre-sented. Another scene from the column'5, which repre-sents the Roman army marching down a short in-clined bridge, has been interpreted by Cichorius as theDacian end of this Danube bridge, as it graduallydescended to the shore. On the basis of the statementby Procopius already mentioned that Apollodorus con-structed an island in the river bed, together wN-ithn-other in a letter written by the younger Pliny (8.4) toCaninius, which speaks of 'rivers turned ilnto newchannels', the German scholar believes that a canalwas laid out at this end of the bridge to take off aportion of the water and thus to reduce the river'scurrent. It is this canal, then, which he finds repre-sented on this second scene, and from it he has recon-structed his idea of the bridgel". Of course there is notrace of any such canal at present, and most investiga-tors believe that Pliny's statement refers merely to Tra-jan's diverting of the course of the Sargetia (Strehl), onwhich Sarmizegethusa, the Dacian capital, was situated,in order to find the treasure of Decebalus, which was

said to be hidden in the bed'". On the reverse of a largebronze medallion in the Cabinet des -Medailles in Paris,one large arch of the bridge is shown"8, and severalCoins of Trajan show fifteen arches of the bridge,and the statue-crowned towers at either end".

The bridge seems to have been built between theFirst and Second Dacian Wars, being completed byIO5 A. D., in good time to transport troops across theriver at the outbreak of the second war. While Diostates that Trajan built it for easy access into Dacia incase of fututre trouble, he adds that Hadrian removedthe surface work because of anxiety lest the barbar-ians might overpower the guards and cross into theprovince of Mfoesia. Hadrian is known to have con-templated the abandonment of Dacia, and to have beendeterred from his project only by anxiety for the Romansettlers there. However, it seems inconsistent that heshould retain the province and yet cut off the only surecommunication with it. So we are led to look for an-other motive than the one assigned by Dio for destroy-ing the bridge. This can readily be found in Hadrian'scharacter. Dio tells how on one occasion, whenApollodorus was showing Trajan the plans of thebridge, Hadrian offered suggestions only to be re-buffed by the architect. This insult may well have ran-kled in the future Emperor's mind, along nith a feelingof jealousy at the talents of Apollodorus, and both ofthese reasons may have led Hadrian not only to destroythe architect, but the bridge as well. Two centurieslater, when, in 328, Constantine was making an ex-pedition against the Goths, the arches and roadwaywere restored, as we learn from the seventh centuryChronicon Paschale (page 527). But the later Gothicand Hun invasions made it necessary finally to dis-mantle it.

Let us briefly see what use Trajan made of the roadand bridge in his two wars against the Dacians.

The Dacians, long a source of anxiety to Rome,had become by Trajan's time a real menace. Theythen occupied lands on either side of the great prom-ontory which the Eastern Carpathian, or Transylva-nian Alps, project upon the Danube, the valley of theTemes (Tibiscus), or the present Hungarian Banat,to the west, and little Wallachia, or western Rumania,on the east. But their real strength was centered in thenorth in the valley of the ,Marosch (Marisia), thechief eastern tributary of the Theiss, in the greateastern loop formed by the Carpathians, a region nowcoextensive with Transylvania in eastern Hungary.It was into this region, well guarded by mountains andforests, that the Romans had to penetrate. There werethree passes which led into the neighborhood of Sarmi-zegethusa (the modern Varhely on the Strehl)-one from the west known as the Iron Gate (2,I52feet), which is not to be confused with the Danuberapids already discussed, and two from Wallachia, theVulcan at the head waters of the Rabon (Schyl), andthe Roterturm, further east, at the head of the Aluta(Alt).

appeared in i872-I874). Duruy gives i6 a-nd Paget 20 (includinlgthe end ones).12See the article Ueber Trajan's Stei-ierne Donaubruecke, inMittheilungen der Zentralcommission zur Erforschung und Erhalt-ung von Baudenkmaelern (Auguist, 1858).13 Page I32, note 5. Duruv gives the spans of the arches as120 feet, or I77 feet from their axes, which makes a bridge of 3,720feet in length, including the end buttresses; Paget gives 3,900feet; Merivale follows Dio's length of 4,770 Greek feet, i. e. 4630.87English feet.14Cichorius, Die Reliefs der Trajanssaule, Tafelband II, P1.LXXII, Nos. 258-26I; Textband III, 13,5 f. (I900).5Cichorius, Tafelband II, P1. LXXIV; Textband III, I53 f.BFig. I4, on page i59.

17SeeDio, 68.14.18Reproducedin Duruy, V, Pt. I, in the figure on page 238."RH.Cohen, Description Historique des Monnaies Frapp6es sousl'Empire Romaine, II, Nos, 542-544 (1882).


6/7

DECEMBER 8, 1924] THE CLASSICAL WEEKLY 63The Dacian Wars, as is well known, furnish in-

soluble problems. If our knowledge of Trajan's reignis meager, that of these wars, its chief featture, isespecially so. Even the causes of the first war arenot clear, whether they are to be laid to the aggressiveexpansion of the military state of Decebalus, or per-haps to the foreign policy of Trajan, which looked bothto the safety of the Danube frontier and to the Em-peror's own glory. In any case, the Romans easilyfound a pretext in irritation at the continuance ofDomitian's tribute, and in the alliance between Daciaand Parthia. Only one fragment remains of Trajan'snarrative of these wars, which we owe to the casualcitation of the grammarian Priscian. It seems to havebeen illustrated, the sketches doubtless forming thebasis of the reliefs on Trajan's column. As literaryevidence is sadly wanting, the only continuous historybeing the one preserved in the late abridgement of Dioalready mentioned, it is to these reliefs that historiansmust turn for their chief knowledge of these wars.While localities and the sequence of events as shown bythem are largely conjectural, still many details of thetwo campaigns can here be read. For we see the as-sembling of Roman armies, encampments, crossings ofrivers, battles, sieges, buildings, landscapes, and bothRoman and Dacian arms and costumes. Duruy hasrightly said that this spiral band of reliefs, which is over6oo feet in length and contains over 2,500 humanfigures, is 'for the military life of the Romans whatPompeii is for their civil life', for this epic in stone is anaccurate mirror of scenes which transpired over I,8ooyears ago. Froehner and Cichorius have succeeded inmaking these events live again before our eyes, and inbringing them into a tolerably sequential narrative,from the crossings of the Danube at the beginning of thefirst war to the suicide of Decebalus at the close of thesecond. Modern scholars have done their best topiece together the chief events of the war from thesereliefs, but naturally have come to divergent con-clusions. Perhaps von Domaszewski has been mostsuccessful in sketching their general outline20.

Trajan left Rome in March, IOI, with part of hispretorian guard to meet his legionary force already con-centrated on the middle Danube near Viminacium(Kostolatz in Serbia), the headquarters in UpperMoesia of the seventh legion (Claudia). Here tworoutes were available, one to the north a short dis-tance to Lederata on the river, the other east along theSerbian bank. Trajan in person chose the former, and,having crossed the Danube by a pontoon bridge to themouth of the Apus (Ipek) near the modern hamlet ofUj-Palanka, marched northward to Tibiscum at theconference of the Tibiscus (Temes) and Bistra (Bisztra).The stations along the route are marked on the TabulaPeutingeriana, and, curiously enough, the few wordspreserved from Trajan's lost narrative, the Dacia,prove that this was the route of advance. Anotherarmy under Laberius Maximus, governor of LowerMoesia, took the eastern route, reaching the Danube at

Cuppae (Golubacs), and hence apparently followed therock-hewn road described above, via Taliata (Golu-binje?) to a point opposite Tsierna (Orsova), where acrossing was made by a second bridge of boats. Thencehe followed the valley of the Tsierna (Cserna) north-wards by way of Ad Mediam (Mehadja), finally cross-ing the lower Carpathians westward by the TergovaPass and reaching the Tibiscus near the present townof Karansebes just below the ancient Tibiscum.From Tibiscum the two armies advanced up thegorge of the Bistra toward the Iron Gate, the key to theMarisia (Marosch) valley, and overlooking the Daciancapital. After fruitless overtures of peace by Dece-balus, a battle, says Dio, was fought by Trajan, whowith Laberius seems to have forced the Iron Gate, atTapae, a place not yet identified. In this conflict theRomans, though victors, appear to have sufferedseverely. The reliefs show Trajan visiting the woundedand using his own garments for bandages. The cam-paigning season now being at an end, Trajan found hehad wholly underrated Decebalus's powers of resist-ance, and so, leaving a guard behind, he withdrew toDrobetae. Here, in the following winter, he learnedthat the Dacians, with the help of the Sarmatae and theRoxolani horsemen, had crossed the Danube near itsmouth into the Dobrudja. He hurried down the riverby boat to Oesctis and turned south into Moesia toprevent the enemy from crossing the Balkans intoThrace, while it appears that other Roman ti oopsfollowed the right bank of the river to cut off the bar-barians' retreat. Here they surprised the enemy in anight attack, possibly at the spot where later waserected the mysterious monument known as Tro-paeum Trajani. In any case the decisive engagementwas fought by Trajan himself further south, where thetown of Nicopolis ad Istrum (Nicup in Bulgaria) wasbuilt on the site of the Roman camp. The next springthe final advance into Transylvania was begun,Trajan wisely throwing a network of defences betweenthe Carpathians and the Aluta, and advancing hiscolumns by several routes. He himself seems to havefollowed Laberius's route to the Iron Gate, whileLusius Quietus, his Moorish master of horse, startingperhaps from Bononia (Viddin in northwest Bulgaria),crossed the Danube to Drobetae. From here he pro-ceeded in a northeasterly direction across the valley ofthe Rabon, reaching the river a little below the VulcanPass; thence he marched further east to the Aluta,and northward through the Roterturm to Cedoniae(Hermannstadt) on the north side of the mountains.From here he seems to have made a great easternsweep, finally approaching Sarmizegethusa via Apulumfrom the north. Again Decebalus sued for peace, butTrajan with the help of Lusius took his capital. Thefirst war then ended with Decebalus acknowledging thesuzerainty of Rome and with a Roman garrison left inhis capital city. Trajan left for Rome, where in the fallhe celebrated a triumph and was given the title ofDacicus - one which was yet to be borne by fourEmperors in the next troubled century, by Maximinand his son Maximus, by Decius, and by Aurelian.

The events of the second war are in even greater20Geschichte der Romischen Kaiser, 2. 174 f. (I909). See alsoFrancke, Geschichte des Trajans, I92 f.; H. Schiller, Geschichteder Romischen Kaiserzeit, I, I, 550 f. (I883); and the routes onKiepert's maps at the end of Volumes I and 2 of Cichorius.


7/7

64 THE CLASSICAL WEEKLY [VOL.XVIII, No. 8, WHOLENo. 485obscurity than those of the first. The peace concludedwith Decebalus could not last, and soon the Dacianking was again gathering arms, repairing orts, makingalliances, and, above all, harassing his. northwesternneighbors, the Iazyges, who were now allies of Rome.War was declared by the Senate in I04, and early thenext summer Trajan,his bridge now completed,set outby sea from Ancona by way of Corinth,the Aegean,andthe mouth of the Hebrus in Thrace, finally resolved toend the Dacian menaceforever. On reaching Nicopolisby the military highway built by Augustus he learnedthat the Dacians had destroyed his works in LittleWallachia. Dio laconically says that the Emperor,having crossed the bridge to Drobetae, 'eventually,after a hard struggle, vanquished the Dacians'. Theskillful artist of the reliefs showsthe convergentmarchof the Roman columns on Tapae, and, in the finaladvance of io6, closingin on the capital. Some of thedefenders are shown setting fire to the city, whileothers have retired to the palace to take poison, andstill others have fled. Among the latter is Decebalus,who is pursued northwards into the fastnesses of theMarisia valley, where, to avoid capture when he isfinally surrounded,he commits suicide. So the secondwar ended with the death of the king, the conquest ofhis kingdom, and its conversion into the Roman pro-vinces of Upper (Transylvania)and Lower(Wallachia)Dacia, the former under an imperial legate, the latterunder a procurator,and the thirteenth legion (Gemina)was left as a permanentgarrison.

To repair the ravagesof war, colonists were collectedfrom Dalmatia, Asia Minor, Syria, and further east-ex totoorbe Romano,as Eutropius (8.3) says. Wherebefore there had been one town, now there were four,Apulum (Karlsburg), Napoca (Klausenburg?), andUlpia Trajana Dacica (the reconstructed Sarmizege-thusa) in Dacia Superior, and Tsierna (Orsova) inDacia Inferior. The new provinces were quickly andlastingly Romanized,asthepresentRumanian anguage,so far separated from its Latin sisters in the west,abundantly testifies. Trajanalso built severalmilitaryroads to connect the chief towns, and the great ViaTrajana, which led from the Danube northwardalongthe Aluta and over the Carpathiansby the RoterturmPass, and thence through the Marisia valley acrosseastern Hungary. This road was to be the laterRoman military approach from the west. The olderrock-hewnroad along the Danube leading to the bridgeat Egeta seems thereafter to have been little used.And, as wehave seen, the dismantlingof the bridgewassoon to follow. In fact, both the road and the bridgehad done their work in conquering Dacia. Trajan'sclaim to military glory had been amply proved. Buthis conquests were not to be lasting. UnderGallienus,in 256 A. D., the Goths crossed the Carpathians, anddrove the Romans out except from a few fortresses be-tween the Temes and the Danube. No Roman coin orinscription of a later date has been found in the Dacianregion. By the time of Aurelian (270-275 A. D.)even these survivingtroops were withdrawn, and were

settled in Moesia in the later colony called DaciaAureliana,which, like the earlier Dacia, was subdividedinto two parts, Dacia Ripensis, with Ratiavia as itscapital, and Dacia Mediterranea, with Sardica (Sofia)as its official center.UNIVERSITY oF PENNSYLVANIA WALTER WOODBURN HYDE

A MODERN HYPOCAUSTThe hypocaust, so familiar from the numerous re-mains of Roman baths, seems to have been little usedfor the general heating of buildings-though passagesin Pliny's and Seneca's letfers suggest that sometimeshot air was conveyed through hollow floors and wallsfrom the bath-room to other apartments-, untilRoman houses began to rise in the inclement climateofGermany and Britain. Not till our own time, appar-ently, has this ancient Roman method of heating beenrevived. The magnificent new cathedral in Liverpool

is destined, when completed, to be third among thecathedralsof the world, and in beauty and originalityofdesign already ranks among the finest of Churchbuildings. The work of building, begun in I904, hasprogressed o far that on July I9, I924, the cathedralwasconsecrated with imposing ceremonies. The finishedportion includes the eastern transepts and the centralspace between, the choir, and behind it the ambulatoryand the vestries, with Lady Chapel and chapter-houseon either side. The problem of heating has been solvedin a new way, and part of the description given in theOfficialHandbook should be of interest to students ofarchaeology:In the year I920 the Engineers. . .developed andpatented a system of heating by means of a warmfloor. . . .The method of heating a building by warm-ing the floor was used by the Romans, as may be seeninthe baths at Bath and in the Baths of Caracalla atRome. Here the hot gases and products of combustionfrom a fire outside the room passed under the floor ontheir way to a chimney stack. In the Cathedralalmostthe entire floor space consists of a double floor, en-closinga system of shallow ducts. The warmedairfromstoves. . .is circulated through these dutcts by anelectrically driven centrifugal fan. . . Under thisarrangementhe floorof the Cathedral tself becomesanimmenseradiator,the surfaceof which is so largecom-paredto that of hot water radiation that the tempera-ture requiredat its surfaceis considerablybelow blood

temperature, and for this reason the floor does notfeel warm to the feet. The tempering of the marblefloor will, however, add to the comfort of the congre-gation, and, the heating surface being at a very lowtemperature,there will be no drying of the air. Theheat is evenly distributed and down draughtsare pre-vented because the heating surface is the whole floorarea of the choir, aisles, and transepts....The heat-inc chamber s situated underthe centre of the Cathe-dral". The floorheating apparatus,hot water boilers,air circulatingand ventilating fans are all arranged nthe chamber,and provisionis made for the apparatusthat will be requiredfor heating the nave when thatpart of the Cathedralis built, the ducts being carriedthiough the temporarywall ready for extension.This is surely an interesting instance of survival, or

revival,of Romancivilizationin Britain!HOBART COLLEGE H. H. YEAMES

walter woodburn hyde - 1924 - traians danube road and bridge

Documents