If my own interpretation is correct, this is simply what is left of a jetty; I explain this at some length below.
However, at the time of writing (in 2010), large-scale OS mapping labelled this feature "remains of old causeway", presumably on the basis of the archaeological report (from 1974) that is linked from the end-note. That interpretation found its way onto the OS map. By 2019, their maps no longer annotated the remains as those of a causeway; though not outright endorsement of my own view, this at least indicates some uncertainty. Nevertheless, I have retained the title "Supposed remains of a Roman causeway" for historical reasons; for example, some readers will own older OS maps with the "causeway" marked on them; my keeping the original title of this page will help any interested readers to come across my comments.
- - • - -
The first-edition OS map (c.1858) shows a jetty here, extending from the head of the beach towards
NS4273 : The Lang Dyke, the wall-like feature that is visible in the middle distance, and which was built in 1773. The large green beacon on the dyke is
NS4273 : The Longhaugh Light, and the large hill in the background is Dumbuck Hill, with its large quarry (
NS4274 : Dumbuck Quarry); compare
NS4274 : Dumbuck Hill, Dunbartonshire. Some
NS4273 : Bollards on the River Clyde can be seen on the right.
This photograph is a view along the line of the jetty, and it seems probable that most of what is now visible here dates from that period (for example, compare the cobbles visible in the foreground, which are shown in detail at
NS4273 : Cobbles on site of supposed Roman Causeway, with
NS4672 : Remains of the south jetty).
However, according to a WoSAS report (see the end-note for the link), this had earlier been the course of a Roman causeway across the River Clyde (for my own reservations about this, see below). Before measures were taken to deepen the river channel for shipping, it was possible, here, at the part of the river adjacent to Dumbuck, to walk across the Clyde at low tide (the crossing was generally referred to as "Dumbuck ford").
For part of the "gravelly mound" that is referred to in the WoSAS report as underlying the cobbled way, see the lower-right quadrant of the image. For a view in the opposite direction, from a point partway along the jetty, see
NS4273 : Supposed remains of a Roman causeway.
Naturally, no signs of passage would be preserved in the river itself. However, the WoSAS report points out that the line of the causeway shown here "aligns with a recognizable road mound on the north bank"; the latter mound is not now very noticeable, but, for what it is worth, the site is shown and discussed in
NS4274 : Linear mound near Milton Island.
- - • - -
In the rest of this item, I will add a few observations of my own on the visible remains.
First of all, the WoSAS report (see the end-notes for the link) cites articles published in the journal "Discovery and Excavation in Scotland" in 1974. It seems odd that these DES articles make no specific mention of the jetty that is indicated on the first-edition OS map (c.1860), although it is perhaps implied, since they do make a distinction between the cobbles and an underlying gravelly mound (the wording suggests that the claim for antiquity applies only to the mound, and not to the overlying cobbles). However, I believe that both the cobbles and the gravelly mound can be explained as the remains of a jetty (the course of the "jetty" marked on the first-edition OS map definitely corresponds to the remains shown in this photograph). Before the Lang Dyke was built, numerous lateral jetties were constructed from the shores of the Clyde; all were built with the same aim, namely, of narrowing the channel to increase the rate of flow.
These jetties are known to have been earth and gravel mounds with a stone covering: the book "Clyde Navigation", cited in
NS4273 : The Lang Dyke, says that "stone quarries were opened up to provide solid material for covering the jetties, borrow pits were created to supply the earth and gravel for the centres, and trees were felled to provide timber for supports".
This method of construction accounts for both the cobbles (the coverings) and the underlying gravelly mound (the centres). If the remains in this photograph are those of a jetty, then they may have been better preserved than other lateral jetties on account of being protected by the Lang Dyke. Later on, the jetty would have been useful as a means of approach to
NS4273 : The Longhaugh Light, as the present picture shows.
The two parallel lines of seaweed that start not far ahead in the present picture might well correspond to a double line of timber posts, for example, that supported a jetty.
What, though, of the mound that continues in the opposite direction (south), through grassland, beyond the beach? It is perhaps worth noting that a couple of jetties, nor far to the east, show continuations (clearly visible on satellite imagery) to the south, through grassland, petering out only where they have been obliterated by a modern golf course; see
NS4473 : Remains of a jetty and
NS4473 : Remains of a jetty. Speaking of the many lateral jetties that had recently been built from the shores of the Clyde, John Golborne, the engineer responsible for the Lang Dyke, noted in 1781 that "the spaces between many of the jetties were filled up and covered with grass, to the great emolument of the proprietors and advantage of the river". These grassy continuations may therefore simply be parts of the jetties, in areas that are now grassland but which were formerly on the shore.
Comparison of these grassy continuations with the mound that leads south from the remains shown in this photograph might establish whether the method of construction is the same, or different.
[The report of the "causeway" is one of a series that were made in the 1970s, and in 1980, and the authors linked that causeway to a road over Dumbarton Muir; however, the related reports describing the course of that road are themselves problematic, though for altogether different reasons: see
Link where I discuss those issues at much greater length.]
There is no doubt that the Clyde could be crossed near here in centuries past. However, although I am not myself an archaeologist, and so try to avoid being overly dogmatic in the views I express here, I believe that it would helpful if the report of a Roman causeway at this site was re-assessed.
[2019 update: the OS map has indeed been updated, as was mentioned at the start of this item.]
This is not the only interpretation of an antiquity in this area that I have questioned. For example, a "crannog" beside the Erskine Bridge was merely a jetty leading to a rock that supported a beacon:
NS4672 : Remains of the south jetty. That site is not to be confused with the undoubtedly genuine
NS4572 : Erskine Crannog nearby.