Why Did Sumerian Scribes Calculate Backwards? The Reverse Method for Resource Allocation
In 2600 BCE, in the city of Shuruppak along the Euphrates, a scribe named Gal-sal faced an impossible task. The temple complex needed to allocate grain stores among competing priorities: seed for planting, rations for workers, offerings to the gods, and emergency reserves. One decision could mean prosperity. Another, starvation.
We know from recovered clay tablets that Sumerian scribes didn't start with what they had and divide it up. They did something counterintuitive: they calculated backwards from required outcomes. They'd determine the minimum grain needed to prevent spring famine, work back to the rations required per worker per day, then reverse-engineer how much must remain after offerings. Only then would they look at current stores and make cuts—starting with the least critical category and moving upward.
This wasn't theoretical. The tablets from the Sumerian Early Dynastic period show actual calculations using this reverse method, called "nig-kas" (literally "account from the end"). One administrative text from Girsu lists harvest allocations where the scribe literally worked from bottom to top of the tablet, starting with "what must remain" before noting "what we possess."
Most modern resource allocation starts with the budget. We open the spreadsheet, see our time or money or headcount, and ask "how should we divide this?" This forward calculation feels rational, but it systematically underweights critical minimums. When you start with resources, you're always tempted to fully allocate them. The team has 40 hours per week, so we schedule 40 hours of meetings and tasks. The department has $200K, so we find ways to spend $200K.
The Sumerian method reveals what forward calculation hides: some resources aren't divisible without breaking the system. You cannot allocate "most" of the grain needed for seed and expect "most" of a harvest. You cannot give your senior engineer 70% of the uninterrupted time she needs for architecture work and expect 70% of a good system design. Certain minimums are categorical, not proportional.
When Sumerian scribes calculated backwards, they were forced to confront a harder question: "What can we absolutely not do?" rather than "How can we do everything?" The nig-kas method made subtraction visible and addition difficult. You couldn't simply add another temple project without explicitly removing something from the bottom of your calculation.
Modern teams practice a kind of resource optimism. We believe we can fit one more initiative, one more responsibility, one more meeting. We're doing forward calculation: "We have these resources, surely we can accommodate this too." The Sumerian scribe would have recognized this as the thinking that leads to famine.
Consider how this applies to quarterly planning. The forward method asks: "We have three teams and twelve weeks, how should we divide this capacity?" The backward method asks: "For our existing committed projects not to fail catastrophically, what minimum capacity must remain untouched? Given this, what new capacity actually exists?"
The difference is stark. Forward calculation makes everything seem negotiable. Backward calculation reveals non-negotiables. It forces you to define what "failure" means concretely before you start distributing resources. The Sumerians knew that "insufficient seed grain" wasn't a 10% problem—it was a civilization-ending problem.
This matters because modern work has many invisible minimums that function like seed grain. The thinking time required for architectural decisions. The slack needed for urgent customer issues. The recovery time between intense projects. When we forward-allocate resources, these minimums become the first casualties because they're not scheduled tasks—they're capacities.
The Nig-Kas Practice:
This week, choose one resource you control: your time, your team's capacity, or your budget. Before planning how to allocate it, list three outcomes that would constitute categorical failure if they occurred. For each, calculate backwards: what minimum resource must be reserved to prevent this? Subtract these minimums from your total. Only now, with what remains, begin allocation. If nothing remains, you haven't discovered a resource problem—you've discovered a commitment problem. Like the Sumerian scribe, you must now make visible which obligation you'll abandon rather than pretending all obligations can coexist.
We know from recovered clay tablets that Sumerian scribes didn't start with what they had and divide it up. They did something counterintuitive: they calculated backwards from required outcomes. They'd determine the minimum grain needed to prevent spring famine, work back to the rations required per worker per day, then reverse-engineer how much must remain after offerings. Only then would they look at current stores and make cuts—starting with the least critical category and moving upward.
This wasn't theoretical. The tablets from the Sumerian Early Dynastic period show actual calculations using this reverse method, called "nig-kas" (literally "account from the end"). One administrative text from Girsu lists harvest allocations where the scribe literally worked from bottom to top of the tablet, starting with "what must remain" before noting "what we possess."
Most modern resource allocation starts with the budget. We open the spreadsheet, see our time or money or headcount, and ask "how should we divide this?" This forward calculation feels rational, but it systematically underweights critical minimums. When you start with resources, you're always tempted to fully allocate them. The team has 40 hours per week, so we schedule 40 hours of meetings and tasks. The department has $200K, so we find ways to spend $200K.
The Sumerian method reveals what forward calculation hides: some resources aren't divisible without breaking the system. You cannot allocate "most" of the grain needed for seed and expect "most" of a harvest. You cannot give your senior engineer 70% of the uninterrupted time she needs for architecture work and expect 70% of a good system design. Certain minimums are categorical, not proportional.
When Sumerian scribes calculated backwards, they were forced to confront a harder question: "What can we absolutely not do?" rather than "How can we do everything?" The nig-kas method made subtraction visible and addition difficult. You couldn't simply add another temple project without explicitly removing something from the bottom of your calculation.
Modern teams practice a kind of resource optimism. We believe we can fit one more initiative, one more responsibility, one more meeting. We're doing forward calculation: "We have these resources, surely we can accommodate this too." The Sumerian scribe would have recognized this as the thinking that leads to famine.
Consider how this applies to quarterly planning. The forward method asks: "We have three teams and twelve weeks, how should we divide this capacity?" The backward method asks: "For our existing committed projects not to fail catastrophically, what minimum capacity must remain untouched? Given this, what new capacity actually exists?"
The difference is stark. Forward calculation makes everything seem negotiable. Backward calculation reveals non-negotiables. It forces you to define what "failure" means concretely before you start distributing resources. The Sumerians knew that "insufficient seed grain" wasn't a 10% problem—it was a civilization-ending problem.
This matters because modern work has many invisible minimums that function like seed grain. The thinking time required for architectural decisions. The slack needed for urgent customer issues. The recovery time between intense projects. When we forward-allocate resources, these minimums become the first casualties because they're not scheduled tasks—they're capacities.
The Nig-Kas Practice:
This week, choose one resource you control: your time, your team's capacity, or your budget. Before planning how to allocate it, list three outcomes that would constitute categorical failure if they occurred. For each, calculate backwards: what minimum resource must be reserved to prevent this? Subtract these minimums from your total. Only now, with what remains, begin allocation. If nothing remains, you haven't discovered a resource problem—you've discovered a commitment problem. Like the Sumerian scribe, you must now make visible which obligation you'll abandon rather than pretending all obligations can coexist.