HistoryPorn

Knitting is generally accepted to have been invented in Egypt around the Millennium, with the first clearly knit object (intricate colorwork socks) dated to around 1100 CE. Roman artifacts from 2-4th centuries would not be “knitting” tools, since knitting as we know it didn’t exist yet. Nalbinding predates knitting, but requires a flat needle with an eye which is fully passed through each loop, so this tool is still not related. It could be some kind of weaving/textile tool, but those knobs don’t actually look that easy to work with. I would expect some kind of taper to facilitate passing threads/yarns over the tip if that were the case. My opinion: it’s a coincidence that it just happens to look like the “knitting looms” at the hobby store.

There are similar objects that are icosahedral (d20 - shaped) that do not have holes, but do have the external nubs.

The glove theory is popular, but that style of knitting isn't known to have been practiced for more than a thousand years after these objects were created. They have never been found with knitting needles, or other thread-related tools.

There's also a bit of a paradox with record keeping. Very common things were almost never recorded - what would be the point of that after all, everyone already knows about them - and these objects don't appear in surviving records. However, we haven't found enough of these for them to have been all that common, and they seem to be confined to Gallo-roman areas, suggesting a sub-cultural difference.

The unsatisfying answer is that we will likely never know.

There have been several other articles about these in the past few weeks.

One of the most common suggestions is exactly that, used for knitting fingers for gloves.

Another I saw that made sense was for securing tent poles, with the nubs being used to secure the canvas to the ground.