Sigma & Diglore | Character dialogue

Diglore

I’ve been mapping the logistics of the Maya’s irrigation systems—think of it as a pre‑industrial supply chain optimization. How do you quantify the ROI of such ancient infrastructures?

Sigma

ROI for ancient irrigation is a simple equation: net benefit divided by total input. In the Maya case, measure water throughput in cubic meters per day, then translate that into agricultural output—tons of maize or beans per hectare. Convert that yield into market value based on current commodity prices, then subtract the capital cost: labor hours, stone, stone quarrying, labor, time lost to irrigation. Finally, divide by the number of years the system remained functional. That gives you a yield‑per‑investment ratio. If the system ran for 200 years and delivered an average of 3,000 tonnes of maize a year at $1.50 per tonne, that’s $4.5 million in gross revenue. Subtract construction and maintenance—say $300,000—and you’re left with $4.2 million. Divide by the original investment of $200,000 and you get a 21‑fold return. That's the ROI, expressed in a metric you can compare against any modern project. If the numbers don't line up, the system is inefficient and needs a redesign—no room for idle speculation.

Diglore

That’s the textbook approach, but you’re treating a 12th‑century hydraulic network like a 21st‑century corporation. First, you’re assuming a static commodity price of $1.50 per tonne of maize—yet the Maya were trading cacao, jade, and other goods, not selling maize on a global market. Second, you’ve lumped “capital cost” into a single figure of $300,000; that’s a modern estimate of labor and stone. What were the actual labor days, the skill level required, the seasonal constraints? And the 200‑year lifetime you’re quoting—did the system actually run that long without major overhauls? If it only operated effectively for 50 years, the ROI shrinks dramatically. I’d start by reconstructing the actual throughput, the seasonal variation, and the social cost of labor before we talk about dollars. The math looks neat on paper, but the variables are a mess in the real world.

Sigma

You’re right the numbers I threw out are rough, but that’s how you start. First, get the raw throughput: the volume of water moved each season, the acreage irrigated, the yield per acre. Archaeologists can estimate that from the size of the canals, the slope, and the crop residues. Next, translate the yield into a value that matters to the Maya: cacao pods per hectare, jade mining output, trade volume. Use relative ratios, not dollar prices. For labor, record the total person‑days: a stone cutter’s day is different from a farmer’s, so weight them by skill level. Seasonal constraints show up as downtime, so factor that into the effective operating days. Finally, instead of a single lifetime figure, map the maintenance cycles you can see in the layers of sediment or repairs in the stone. Then you’ll have a real ROI that accounts for all those variables. The trick is to keep the data granular—no one likes a one‑size‑fits‑all estimate.

Diglore

Nice, you’ve pulled the weeds out of the rough numbers. But keep in mind the sediment layers you’ll use to date repairs are themselves a patchwork of flood events and droughts—sometimes what looks like a maintenance cycle is just a natural scar. Also, weighting a stone cutter’s day against a farmer’s is tricky; their labor isn’t interchangeable, but their outputs overlap. Maybe start with a single, well‑documented canal and run the numbers through your model; if the ROI spikes or plummets, that’s a red flag. The devil’s in the details, as always.

Sigma

Exactly, you can’t just cherry‑pick data. Start with one canal that has a complete stratigraphic record: the exact dates of construction, each repair, and the seasonal water flows. Map each repair to a specific event—flood, drought, or intentional upgrade—using isotopic markers. Then assign labor units: a stone cutter’s day is a high‑skill, high‑output unit, a farmer’s day is low‑skill, high‑volume. Normalize them to a common metric like “person‑hours of high‑skill labor.” Run the ROI calculation with those precise inputs. If you see a jump in ROI when a particular repair is added, that’s your red flag. If the numbers stay flat, the canal was efficient from the start. Either way, you’ll know whether the Maya were beating the market or just surviving.

Diglore

Sounds solid, but remember those isotopic markers can be ambiguous—what’s a flood signature versus a drought signature in the same strata? I’d run a sensitivity test on the repair dates; if shifting a single event changes the ROI by, say, ten percent, that’s your red flag. Also, don’t forget the social context: a “high‑skill” stone cutter might have worked in short bursts, whereas farmers could spread labor over longer seasons. If the ROI stays flat across those variations, we can claim the Maya engineered efficiency. If it spikes, maybe they were chasing profit. Either way, the data will tell.

Sigma

Run the test, shift each repair by a year, watch the ROI swing. If it jumps ten percent, that’s a flag. Keep the labor weights fixed but adjust the duration of stone‑cutting bursts versus farmers’ spread. If the curve stays flat, the Maya nailed it. If it spikes, they were chasing gains. In either case, the numbers will lay the groundwork. Keep the data tight, the model lean, and you’ll avoid the pitfalls of vague markers.