River barge from Lipe on the Ljubljana moor

The Ljubljana basin is an exceptionally passable region between the Danubian area, Italy and the Adriatic. Consequently, it represents the junction of trade and traffic flow in the south-eastern Alpine region from old. Various excerpts from Roman literature point to the significance of rivers for trade and traffic during the pre-Roman era. The large river barge discovered in 1890 at Lipe on the Ljubljana moor best confirms these data provided by Roman geographers.



The discovery

The barge was discovered by workers digging a trench on the property of dr. Josip Kozler between the Ljubljanica river and Ljubljana - Podpec road. Presently, the owner reported the discovery to the curator of the Regional Museum, Alfons Müllner, who immediately recognised the significance of the discovery. Excavations were carried out between the 25th and the 31st of October along with documentation, complete with drawings and photographs. Due to poor weather, work efforts continued the following summer. In July 1891, the vessel was measured once again for the woodworker, Karol Binder, to construct a model on a scale of 1:10 and this lead to the creation of another photograph. Finally, numerous wooden parts were collected from two cross-sections and taken to the museum.

The stratigraphic position

The barge lay in the loamy chalk with it's bow facing West and it's stern facing East. There was no trace of any cargo in the relatively well preserved barge, which could suggest that it was abandoned due to it's declining age, deterioration or possibly some other reason. Excavation revealed that it was filled with a 30 cm thick organogenic layer, while the uppermost parts reached into a 45 cm thick layer of peat that covered the organogenic layer and the barge. An additional layer of turf, 3 to 4 m thick, was removed for other purposes already prior to the excavation. On the basis of the stratigraphic results it can be concluded that the barge navigated the meandering currents if the Ljubljanica river, although due to it's shallow draught it was also suitable for navigation across the often flooded plain. For some inconceivable reason the barge was abandoned in a dead branch of the then riverbed or in a flooded region nearby. The former seems more probable considering the fact that rivers channelled their riverbeds in the loamy chalk of the Ljubljana moor, which is where the barge eventually lay. Gradually, the barge sunk to the floor of the riverbed or basin where it was then covered by an organogenic layer.

The chronology

Based upon the fact that the barge was situated in loamy chalk and under a layer of turf, Müllner reasoned that it navigated the water currents during the period that a lake existed and that it sunk before the lake became marshy. He attributed the rendering of the lake to marsh and the formation of peat to the chronological period between 500 and 100 BC, while he associated the barge with Etruscan commerce between Italy and the East during the perlod between 900 and 400 BC. Certain authors adjudged a much younger age to the barge due to its similarity with vessels from the Middle Ages and modern times.

Radiocarbon analysis (performed in Groningen in the Netherlands) of the samples of the planks (Picea sp.; GrN-20813), the stopper (Fraxinus sp.; GrN20812), with which knots in wood were usually substituted, as well as the cross timber (Quercus sp.; GrN-23548), registered that the stopper is 2140 +- 20 BP years old, the planks are 2135 +- 30 BP years and the cross timber is 2090 +- 35 BP years old. Following the calibration method presented by Stuiver and Kraeds (1986), absolute dates spanning from 210 to 150 BC were obtained with 83 % probability for the first two samples, and values spanning from 185 to 90 BC with 68 % probability for the third sample. Thus it would seem that trees intended for building purposes were cut down sometime during the beginning or the middle of the 2nd century BC, while the barge was probably in use during the second half of the 2nd and the beginning of the 1st century BC.

The construction

The National Museum of Slovenia currently preserves 110 wooden parts of barges in its depot. In addition to the preserved parts of the barge, a rather precise description of Müllner's, original sketches from the field and drawings published in Argo, the photographs mentioned above as well as Binder's model are all available for the deliberation concerning the construction of the barge.

The reason for Müllner's meagre documentation drawings lies in the precisely bullt model measuring 3 in in length and exhibiting even the greatest of details. Despite the exceptional precision, some details were somewhat revised on the model (e.g. the chine-girder as well as the joints in the side planks) - probably for the purpose of simplifying construction. Consequently, the descriptions of individual parts of the barge are based upon the preserved parts, photographs, drawings and also the model, considering that there is no reason to doubt the integrity of the woodworker's measurements.

The structural characteristics of the barge from Lipe classify it as belonging to the group of sewn boats. At the same time it represents the oldest example, or rather prototype, of a river barge classified as the so-called "Roman - Celtic" type. The shape of the barge corresponds with the term "pontonium" (Isid. Etym. 19, 1, 24), the Roman expression for a shallow wooden boat with a flat bottom and a small draught and which is intended for transporting cargo. The barge was a symmetrical, long oval shape with truncated ends. It measured approximately 30 in in length, up to 4.8 in in width and 0.6 in in height. It had a flat, keel-less bottom with low sides that curved outward. Floor cross timbers with knees between them which supported the sides, constructed the internal supporting structures. The shell was composed of bottom and side planking made from longitudinally placed planks that were sewn together. The side planks were joined additionally with wooden tenons inserted in notches drilled in the edges of facing planks. The passage from the bottom to the side planks was constituted by slightly curved planks, otherwise referred to as chine-girders. Wooden tenons, or dowels, and a few nalls joined the planks and the internal supporting structure. The similarly constructed bow and stern terminated simply with an inclined wall and they can be differentiated solely by the space situated for the steersman. The barge was moved by being pushed off and it had a gangway timber for this purpose. Two strakes lay along the middle of the barge and they served as additional support to the bottom as well as for the placing of cargo.

The board at the stern probably served as a standing space for the steersman and indicates that a steering oar must have been used to help navigate through the otherwise tranquil currents of the Ljubljanica. Perhaps a board, constructed from one piece of oak wood and with a raised part in which two holes were drilled is also of some relevance. Both holes indicate that they served a particular purpose, possibly something to do with rope manoeuvres (for towing or mooring?).

Xylotomous analyses of the samples indicated that spruce(Picea sp.), oak (Quercus sp.), ash (Fraxinus sp.) and black alder (Rhamnus carthartica) wood were used for the construction of the barge. The shell was constructed using a lightweight yet resinous sprucewood, as opposed to the high quality wood of a fir tree (Abies alba) which can be easily handled; this is somewhat surprising considering that sprucewood was not highly appreciated in Roman times (Plin. 16, 40, 42, 90). Similar throughout the Mediterranean where hard oak wood was used solely for the keel and ribs, the knees and part of the cross timbers on the barge from Lipe were also of hard oak wood. The resistant, flexible and hghtweight wood of ash tree was used for the cross timbers, the gangway and the two stakes. Müllner mentions that marks left by the usage of an axe, saw, file and borer can be traced on the barge. All of the above mentioned tools can be classified as standard tools used by Roman shipbuilders. The delicate construction of the barge certainly required frequent maintenance work and repairs considering that sewn boats, due to loosened cords and deteriorated sealants, need to be dismantled, have their sealants refurbished and then sewn back together again. Inasmuch as the barge from Lipe was regularly used and maintained, it presumably survived a period of 10 to 20 years; it definitely did not last longer than 50 years.

The sequence of construction, or rather, the composition of the individual structural parts indicates that the shape of the barge was conceived as a bottom-based construction. The entire shell was built first and then the internal supporting structures were placed within it. The symmetry of the individual segments of the shell (the distribution of the bottom and side planks and the joints between them), as well as the uniform intervals between the cross timbers and the knees, suggest that geometric criteria were observed by the shipbuilders during construction.

The entire tonnage of the barge was calculated to be approximately 5 tons, based upon the volume and the specific weight of the individual parts. The flat bottom with a work surface of approximately 108 square metres and the low and inclined side planking suggest that the river barge had a carrying capacity of at least 40 tons and a maximum draught of approximately 40 cm.

The structural characteristics (the joining of planks and their symmetrical distribution, the use of dowels and nails, the standardised proportions between the individual elements), the citations in written sources referring to the Late Republican era in the Cisalpine region, as well as the presence of the so-called 'Romano- Celtic' type of vessel in the region north of the Alps only after the Roman conquest, all compose reasonable arguments that the technological origins for the barge from Lipe lie among the Mediterranean vessels of the sewn type. The development of sewn vessels with flat bottoms is linked particularly with the geographic conditions of the northern Italian - Po region. The size and carrying capacity of the barge from Lipe, which, according to current findings, is present in a La Tene environment, suggest that organised river transport of larger quantities of goods was active between the northern Italian and eastern Alpine regions during the 2nd century BC. There are three possible hypotheses concerning the origins of the barge that are grounded upon the execution of political-economical events in the south-eastern Alps at the beginning of the expansion of Roman influence: 1.) The vessel represents a river barge belonging to Italian merchants who, according to written sources, established contacts already very early with tribes residing in the hinterland of Aquileia, 2.) The barge was a transport vessel belonging to native inhabitants (Celtic Tauriscans?) who built the vessel on the basis of a foreign model and then used it for the transport of goods, 3.) The barge served as a military supply vessel and is thus connected with one of the Roman military interventions against the Danube region.

In consideration of all the available data, the hypothesis conjecturing a military supply vessel seems the most likely. It is of considerable significance that all vessels bullt in the Mediterranean tradition (mortise-and-tenon joints) only appear north of the Alps subsequent to the incorporation of these regions into the Roman dominion. The barge from Lipe would thus represent one of the earliest archaeological pieces of evidence confirming Roman intervention towards the East. Presumably, the barge served to transport any and all necessary goods from Vrhnika to Zalog, where the cargo was transferred to smaller vessels more suitable for navigation through the menacing Sava gorge.

Fig 1.: The barge from Lipe, July, 1891. View towards the East (unknown author; archives of the Department of Archaeology at the National Museum of Slovenia, no. 248).

Fig 2.: Alfons Müllner at the conclusion of the excavation of the barge in Lipe, October 31st, 1890. View towards the West (photograph: Gustav Pirc; archives of the Department of Archaeology at the National Museum of Slovenia, no. 253).

The previous text is a shortened version of an article written by Andrej Gaspari, which was published in Arheoloski Vestnik 49, Ljubljana 1998 under a title: A "Pontonium" from Lipe on the Ljubljana moor.


Author` s adress: Andrej Gaspari
Oddelek za arheologijo
Zavetiska 4
1000 Ljubljana
Slovenia