It's archaeology's version of a theory.

You see a huge city that appears to have vanished pretty fast, so you come up with some ideas about how/why that happened, like, say, ecological collapse (hypothesis). Once you find some plausible evidence that for that, you flesh out what that would look like more broadly. Now you have a theory (narrative) which should have some explanatory power. Then, as you encounter other evidence, you expect it to fit with and support that theory (narrative). If it doesn't, it may turn out your theory (narrative) was wrong, or at least incomplete, just as in the rest of science.

Just like theories in other fields, these narratives can be kind of sticky and survive longer than the evidence suggests they should.

It's not a perfect mapping, but that's basically what's going on.

Narratives are stories. In this case a story used to describe a conjecture. Unfortunately researchers found the story convincing enough that Since the publication of the wood‐overuse hypothesis, no attempts have been made to evaluate if erosion in the uplands and/or increased flooding in the floodplain did indeed occur during Cahokia's occupation.

The authors have a potential solution in that conjectures that cross academic divisions should include support from related academic fields or from new interdisciplinary ones like geoarchaeology.

Moving forward, we propose that to move past these older narratives of ecocide, there needs to be increased engagement with obtaining both archaeological and environmental data to address these older theories, a need for researchers who are trained in interdisciplinary research, as well as increased support for long‐term, interdisciplinary collaborations. The interdisciplinary field of geoarchaeology is especially equipped to help us move forward, as long as the members of the field remain engaged with developing anthropological theory.

Narratives are chains of events, which can be used to explain an outcome. Not every narrative is political.

To me narrative have fleshed out stories with unsupported bits and pieces. I’d prefer just stating what is understood at the time (which obviously can be wrong and disproven layer on but at least it’s in good faith).

In these situations, the evidence is rarely clear cut. Part of establishing the belief in the evidence is investigating whether or not it can fit into a plausible-but-ultimately-speculative narrative.

In other words, if you can't even construct a coherent story of causation that incorporates your evidence, then maybe you have misinterpreted your evidence.

That's why a narrative is an important part of the puzzle. They are not to be taken as gospel -- they're just a basic sanity check of the evidence.

Edit: I'm not dismissing your perspective! I would have said the exact same thing a couple of years ago. I'm used to evidence being either obviously true or obviously false, or at least being able to establish a Bayesian posterior belief based on incontrovertible assumptions.

What changed for me was that I read some monographs in anthropological history (or is it historical anthropology? I keep forgetting which is which.) These make for incredibly frustrating reading for someone like me. It's all anecdotes and narratives and no clear sets of underlying principles. But that's precisely why it's been so meaningful for me to read them. It's really widened my perspectives on both humanity and that type of research.

Narratives can be informed more by the time from which they arise and not by the time to which they want to describe. The authors provide the context here:

Environmental explanations for the collapse of complex societies have been popular topics since William Thomas' 1956 volume on “Man's Role in Changing the Face of the Earth” (Thomas, 1956). This seminal work established the philosophical argument that humans are inherently destructive to the environment (Middleton, 2017; Ponting, 1991; Thomas, 1956), a philosophy that is widely applied in anthropology (Oliver‐Smith & Hoffman, 1999), geology (Nianfeng et al., 1999; Wilkinson, 2005), biology (Ceballos et al., 2015; Meyer & Turner, 1992; Vitousek et al., 1997), environmental ethics (Attfield, 2008), and general public discourse today (Diamond, 2005; Goudie, 2019; Montine et al., 1990; Ponting, 1991; Ward, 2009).

I imagine too the narrative, not despite, but because of the "holes" gives researchers new areas to focus their investigation. Then either buttress the narrative by filling in the holes or to challenge it.

A friend was reading a paperback in the 70's that suggested dinosaurs were warm-blooded. I thought it was an insane premise (because I was maybe 12 years old) but there does seem to have been a tidal shift in how dinosaurs are perceived vs. how I was taught as a child. No doubt it was someone with a theory or narrative that gave pause to some of the researchers in the field, kindled the imagination of some young researchers....

Thus the need to fill holes with assumptions built on assumptions influenced by current societal trends, aka a narrative, & package the product as "Science".

Stating what's "understood" in archaeology is just as much an act of interpretation, except you'll be less able to comprehend it as a layperson because it's filled with irrelevant details like "sites 1, and 3, and 9 show tree ring dates of 1210, 1360, and 1297, while sites 1a, 2, and 6 show archaeomagnetic dates of 1500, 1678, and 1349 respectively". Imagine that for dozens of kinds of evidence across thousands of sites. It will still be subject to many of the pitfalls of interpretation in how you recognize a site, label a site, excavate it, date it, and so on, but that's basically the "base level" evidence in archeology.

A narrative can be a hypothesis, a hypothesis for the [hi]story (i.e. sequence of events) of what happened.