Ancient Babylonian astronomers used the area of a four-sided shape to work out how far Jupiter would appear to move across the sky. The calculations survive on clay tablets dated to about 350–50 BCE, and a sketch you can make on scrap paper shows why the method works. That sketch is ours, not a graph the scribes are known to have drawn.
The problem: Jupiter doesn’t move the same amount every day
The tablets deal with Jupiter’s apparent movement along the ecliptic, the path the Sun appears to trace against the stars over a year. In one calculation, the clock starts when Jupiter first becomes visible again as it rises at dawn. Mathieu Ossendrijver, the historian of astronomy who analyzed the tablets, shows how the procedure finds the planet’s change in position over the next 60 days. In the model, Jupiter moves less per day at the end of that stretch than at the beginning. Multiply its starting rate by 60 and you’ll count too much movement; use only its ending rate and you’ll count too little.
Draw the calculation with made-up numbers
Imagine a planet’s daily movement steadily drops from 4 degrees per day to 2 degrees per day over 6 days. A degree measures an angle across the sky, not a distance in miles. These are teaching numbers, not measurements or numbers from the tablets.
- Draw a horizontal line from day 0 to day 6. This is your time axis.
- At day 0, mark a height of 4. At day 6, mark a height of 2. The heights stand for degrees per day.
- Connect those two marks with a straight line. Draw vertical lines down from both marks to the time axis, then shade the four-sided space enclosed by the lines. The shaded shape is a trapezoid.
To get its area, average the two heights and multiply by the width: (4 + 2) ÷ 2 × 6 = 18 degrees. The “per day” in the average rate cancels the days, leaving an angle. Put another way, a steady drop from 4 to 2 adds up to the same movement as traveling at 3 degrees per day the whole time. That shortcut works here because the rate changes evenly in our sketch. You can’t use it for just any change in speed.
Now try the numbers from the tablets
In Ossendrijver’s reconstruction of the first 60 days (Science, p. 483), the initial rate is 0.2 degrees per day and the final rate is about 0.1583 degrees per day, converted here to decimals. Using the exact values, the average is about 0.1792 degrees per day. Multiply the exact average by 60 and the result is 10.75 degrees. Add that to Jupiter’s starting position along the ecliptic to get its predicted position at day 60. The calculation tells you how far to go from the starting point; it doesn’t tell you where to start.
One tablet sets out the calculation using arithmetic. Other tablets describe it in terms of the sides and area of the figure. That distinction matters: Figure 2 on p. 483 of the research paper is a modern graph of the reconstructed model. The tablets are evidence for the procedure, not surviving examples of a graph with modern axes.
There’s a neat twist in the fragments. Some describe a way to estimate when Jupiter would cover half the predicted distance: divide the trapezoid into two pieces intended to have the same area. The study explains that the surviving approximation gives two slightly different areas (Science, p. 484). Look back at your sketch. The left side is taller, so it takes fewer days to accumulate the same area there than on the right. Half the movement happens before half the time has passed. That’s what makes the shape useful: you can see the answer before you do any more arithmetic.
For a different example of how researchers reconstruct the ancient world from what survives, see how archaeologists read the layout of Hattusha.
Discover more from Simply Sound Advice
Subscribe to get the latest posts sent to your email.
