The Moon Clock

"The Neolithic Age" ended with two problems that farming made harder: knowing the time of year, and keeping track of what people owe each other as communities grow. People had been working on the first one long before they farmed.

The first clock people had was the Moon. Its phases repeat about every 29.5 days, and anyone can see and count them. The word month comes from moon.

Twelve lunar months add up to about 354 days, 11 days short of the solar year, so a calendar that counts only moons slips a full month against the seasons about every three years.^[1] To keep months tied to the seasons, people had to check the count against the Sun.

The Moon and the Midwinter Sun

The oldest structure that may have done this is at Warren Field in Aberdeenshire, Scotland. By about 8000 BCE, hunter-gatherers had dug 12 pits in an arc, smaller at the ends and larger in the middle. Gaffney and colleagues (2013) argue that the pits track the lunar months, with their sizes following the phases of the Moon. At that time, a pass in the hills framed the midwinter sunrise, which would have let the people using the pits reset their count of months once a year.^[2]

The authors call the calendar reading hypothetical: trees may have blocked the view, and the alignment could be chance.^[2] If it holds, people were checking a count of moons against the Sun about 4,000 years before farming reached Britain.

The Landscape as Calendar

The Sun sets the length of the year. Its rising point on the horizon moves north and south through the year and turns back at the two solstices. Watching where the Sun comes up against a hill or a stone, from the same spot each time, turns that movement into a calendar.

Ezcurra and colleagues (2022) showed how accurate this can be, using a much later society. On the summit of Mount Tlaloc, on the eastern edge of the Basin of Mexico, a stone causeway lines up with the sunrise on February 23–24, the start of the year in the Mexica (Aztec) calendar. The authors argue that this kept the calendar that set the farming year in step with the Sun.^[3]

In late February the sunrise point moves almost a full Sun's width along the horizon each day, so an alignment on that date picks out a single morning. Near the solstices, the authors note, "sunrise will seem to stand still for some 10 d."^[3] A solstice alignment can show that the year has turned, but it can't pin down the exact day by eye.

Rocks That Remember

Some Neolithic communities built the horizon into a monument. At Nabta Playa, in the desert of southern Egypt, cattle herders came to a seasonal lake after the summer rains, and the site became a regional ceremonial center. Around 4800 BCE they set up a ring of stones about 4 meters across, with gaps between upright slabs that frame a rough north–south line and a line toward the summer solstice sunrise.^[4][5] How precise the solstice line is, and whether it was intended, is still debated.^[6]

At Newgrange in Ireland, built around 3150 BCE, an opening above the entrance called the roof-box lets the midwinter sunrise shine down the 19-meter passage to the floor of the chamber.^[7] Ray (1989) found that the passage lined up with the midwinter sunrise of 5,150 years ago to within a few minutes of arc, which supports the view that the orientation was deliberate. He also placed Newgrange among wider surveys showing that megalith builders marked the southern limits of the Sun and the Moon "albeit approximately," and read that low accuracy as a sign of ritual interest more than of calendar making.^[8]

Knowth, another great passage tomb near Newgrange from the same centuries, is often said to face the equinoxes and to track the Moon, but a full survey of its two passages found no equinox alignment and no link to the Moon's extreme rising and setting points.^[9]

The great sarsen stones of Stonehenge were raised between about 2600 and 2500 BCE around an axis that points to the midsummer sunrise in one direction and the midwinter sunset in the other, and its excavators read the layout as focused on midwinter. The rectangle formed by its four Station Stones has a second axis aimed roughly at the Moon's most southerly rising and most northerly setting.^[10] A recent proposal that its stones encode a calendar of 365.25 days was rejected by other archaeoastronomers. Near the solstices the Sun barely moves along the horizon, so the alignment could not have kept such a calendar accurate.^[11]

Each of these alignments recorded something that took years of watching to get right. Once the stones were set, people who had never made the observations could see the year turn, for as long as the monument stood. That is the ratchet effect from "The Neolithic Age," with stone holding the result.

Built and Used Together

Building these monuments took organized labor on a large scale. At Stonehenge, most of the big sarsen stones came from West Woods, 25 km to the north, and a typical upright weighs about 20 tonnes.^[12] The smaller bluestones were quarried around 3000 BCE in the Preseli Hills of west Wales, about 230 km away.^[13] The six-tonne Altar Stone came from northeast Scotland, at least 700 km away.^[14]

People also gathered at these places at set times of year. Durrington Walls, near Stonehenge, was most likely the village of the people who built the sarsen phase, a group of perhaps several thousand. Pigs were killed there all year, with a large peak in late autumn and winter, and the excavators link those feasts to the work of building.^[15] Isotopes in the bones show that few of the pigs were raised locally, and some probably came from Scotland.^[16]

Göbekli Tepe, in southeastern Turkey, is far older. It was built from the tenth millennium BCE, before farming was established in the region. Its excavators argue that feasts brought together groups from a wide area to quarry, carve and raise its stone pillars.^[17] Later work found that some people also lived there year-round.^[18]

A shared date is what the economist Thomas Schelling called a focal point: a choice people converge on without talking, because each expects the others to expect it.^[19] Michael Chwe argues that public ceremonies turn such a choice into common knowledge, so that everyone knows it and knows that everyone else knows it.^[20] In Mesoamerica, Šprajc and colleagues (2023) argue that the builders of the earliest ceremonial centers needed solar calendars so that people living far away knew "when to go to communal ceremonial complexes without the benefit of communication systems over long distances."^[21]

The feasts paid for the work. Dietler and Herbich (2001) describe the work feast, food and drink given in return for a day of communal labor, as "a fundamental economic practice."^[22] For their regions, these monuments were among the largest gatherings of people and labor before there were towns. Later articles in Part 4 follow how cities took over that role, with writing to keep track of the work.

A Monument as a Data Platform

"What is a Data Platform?" describes a data platform in three layers: infrastructure, information and knowledge work. It also sets one rule: the knowledge work layer uses only the information layer and never has to go back to the raw data. A solstice monument follows the same pattern for a single question.

Getting an alignment right took people who knew how to watch the horizon over many years. Once it was built, nobody needed that skill to see when the year turned. The monument stored the result of skilled observation so that people without the skill could use it. The tools in the rest of this series do the same with other kinds of knowledge work. The monument also solved the problem once for everyone who came, the comparison "How Smart City Investment Enables Regional Economic Growth" makes between Roman aqueducts and a city's data platform.

A monument answers only one question, can't be updated without moving stones, and holds nothing about any particular person or exchange. The psychologist Merlin Donald called stores of knowledge outside the brain external symbolic storage, and dated the first ones to the permanent symbols of the late Ice Age.^[23] Writing became the most flexible of them, which is why "How Writing Allowed Information to Become Infrastructure" treats writing as the first data platform. Monuments were an earlier, single-purpose kind.

The Limits of Stone

Monuments work for knowledge that repeats. Midwinter comes at the same point every year, so a fixed alignment keeps giving the right answer. A farming economy also produces information that changes every season: how much grain is in store, which family works which field, and who owes labor to whom.

Stone can't hold that. An alignment can tell a community when midwinter has come. It can't record that the family in the eastern field owes three days of work to the granary and has paid two.

In Mesopotamia, where the first cities grew, those records moved into clay, and the count of months moved with them. The earliest tablets from Uruk, written at the end of the fourth millennium BCE, already use an administrative year of 12 months of 30 days, close to the Moon's cycle but fixed for accounting. Scribes used it to work out grain rations, and the sign for one day's ration was a bowl. Later records used the same months to count days of labor.^[24] "From Barley to Cuneiform," the next article, follows how that accounting grew into writing.

---

References

1. Doggett, L. E. (1992). Calendars. In P. K. Seidelmann (Ed.), Explanatory Supplement to the Astronomical Almanac (pp. 575–608). University Science Books. NASA reprint [§1.1: tropical year 365.24219 days; synodic month 29.530589 days; twelve synodic months 354.36707 days.]
2. Gaffney, V., Fitch, S., Ramsey, E., Yorston, R., Ch'ng, E., Baldwin, E., et al. (2013). Time and a place: A luni-solar 'time-reckoner' from 8th millennium BC Scotland. Internet Archaeology, 34. doi:10.11141/ia.34.1 [§2: all 12 pits cut "by the early 8th millennium BC"; §5: pit sizes "highly suggestive of the lunar phases," and a pass in the hills "framed the rising of the midwinter sun," allowing "an annual solar recalibration." The authors note that the calendar function remains hypothetical and that the alignment could be coincidence.]
3. Ezcurra, E., Ezcurra, P., & Meissner, B. (2022). Ancient inhabitants of the Basin of Mexico kept an accurate agricultural calendar using sunrise observatories and mountain alignments. Proceedings of the National Academy of Sciences, 119(51), e2215615119. doi:10.1073/pnas.2215615119 [Results, "Sunrise Alignments": the Tlaloc causeway aligns with the sunrise on February 23–24; near the equinox the sunrise point shifts about 25′ a day, against a solar disk of about 31′; near the solstices "sunrise will seem to stand still for some 10 d."]
4. Malville, J. M., Wendorf, F., Mazar, A. A., & Schild, R. (1998). Megaliths and Neolithic astronomy in southern Egypt. Nature, 392(6675), 488–491. doi:10.1038/33131 [the complex "has alignments to cardinal and solstitial directions"]
5. Malville, J. M., Schild, R., Wendorf, F., & Brenmer, R. (2007). Astronomy of Nabta Playa. African Skies/Cieux Africains, 11, 2–7. ADS 2007AfrSk..11....2M [p. 2: herders came "after the summer rains"; Nabta served as "a regional ceremonial centre" during the summer wet season; rainfall was "no more than 100–150 mm." p. 3: the circle is "approximately 4 m across," with an adjacent hearth dated 4800 ± 60 BC. A revised version appears in J. C. Holbrook, J. O. Urama, & R. T. Medupe (Eds.), African Cultural Astronomy (pp. 131–143), Springer, 2008.]
6. Belmonte, J. A. (2010). Ancient Egypt. In C. Ruggles & M. Cotte (Eds.), Heritage Sites of Astronomy and Archaeoastronomy in the Context of the UNESCO World Heritage Convention: A Thematic Study (pp. 117–138). ICOMOS and International Astronomical Union. PDF [p. 118: the astronomical intent at Nabta Playa "is highly controversial"; p. 122 lists its solar and stellar alignments as hypothetical.]
7. Patrick, J. (1974). Midwinter sunrise at Newgrange. Nature, 249(5457), 517–519. doi:10.1038/249517a0 [the light travels "along a 19 m passage"]
8. Ray, T. P. (1989). The winter solstice phenomenon at Newgrange, Ireland: Accident or design? Nature, 337(6205), 343–345. doi:10.1038/337343a0 [construction dated to 3150 ± 100 BC; the passage matches the midwinter sunrise of 5,150 years ago to within about 5′, which "supports the theory that the orientation of Newgrange was deliberate." Closing paragraph: surveys show megalith builders marking the southern limits of the Sun and Moon, "albeit approximately," and "such low accuracies suggest that ancient man's interest in these bodies may have been ritualistic rather than for the purpose of calendar construction." A correction (Nature, 342, 958) credits T. M. O'Brien with the surveys.]
9. Prendergast, F., & Ray, T. (2017). Alignment of the western and eastern passage tombs. In G. Eogan & K. Cleary (Eds.), Excavations at Knowth 6: The Passage Tomb Archaeology of the Great Mound at Knowth (Appendix 2, pp. 263–276). Royal Irish Academy. Open copy [no equinox alignment, and "a lack of any association with the major standstill positions of the moon." The lunar reading often cited for Knowth, Prendergast, K. (2017), Journal of Lithic Studies, 4(4), 67–76, reports no new measurements.]
10. Darvill, T., Marshall, P., Parker Pearson, M., & Wainwright, G. (2012). Stonehenge remodelled. Antiquity, 86(334), 1021–1040. doi:10.1017/S0003598X00048225 [Table 3, p. 1026: Stage 2, 2620–2480 cal BC, with the "SW–NE solstitial axis (midwinter sunset/midsummer sunrise)"; p. 1030: the layout "strongly suggests that its principal focus was the midwinter solstice"; p. 1033: the Station Stones' other axis is "aligned approximately on the major southern moonrise … and the major northern moonset."]
11. Magli, G., & Belmonte, J. A. (2023). Archaeoastronomy and the alleged 'Stonehenge calendar'. Antiquity, 97(393), 745–751. doi:10.15184/aqy.2023.33 [a reply to Darvill, T. (2022). Keeping time at Stonehenge. Antiquity, 96(386), 319–335. The abstract calls the 365.25-day reading "unsubstantiated"; the preprint (arXiv:2211.07981) argues that "the slow movement of the sun at the horizon makes it impossible to control the correct working of the calendar." Darvill replied in Antiquity, 97(393), 752–754.]
12. Nash, D. J., Ciborowski, T. J. R., Ullyott, J. S., Parker Pearson, M., Darvill, T., Greaney, S., et al. (2020). Origins of the sarsen megaliths at Stonehenge. Science Advances, 6(31), eabc0133. doi:10.1126/sciadv.abc0133 [50 of the 52 surviving sarsens share one source, most probably "West Woods, Wiltshire, 25 km north of Stonehenge"; a typical upright weighs about 20 tonnes.]
13. Parker Pearson, M., Pollard, J., Richards, C., Welham, K., Casswell, C., French, C., et al. (2019). Megalith quarries for Stonehenge's bluestones. Antiquity, 93(367), 45–62. doi:10.15184/aqy.2018.111 [p. 45: the Preseli Hills of west Wales, "230km away," with "megalith quarrying around 3000 BC."]
14. Clarke, A. J. I., Kirkland, C. L., Bevins, R. E., Pearce, N. J. G., Glorie, S., & Ixer, R. A. (2024). A Scottish provenance for the Altar Stone of Stonehenge. Nature, 632(8025), 570–575. doi:10.1038/s41586-024-07652-1 [the Orcadian Basin of northeast Scotland; "a 6 tonne shaped block … sourced at least 750 km from its current location."]
15. Craig, O. E., Shillito, L.-M., Albarella, U., Viner-Daniels, S., Chan, B., Cleal, R., et al. (2015). Feeding Stonehenge: Cuisine and consumption at the Late Neolithic site of Durrington Walls. Antiquity, 89(347), 1096–1109. doi:10.15184/aqy.2015.110 [pigs were killed year-round, with "a substantial peak … during the late autumn and winter"; Durrington is "most likely the village of the builders of Stonehenge Stage 2," a group that "probably numbered several thousands"; the feasting "accords well with … voluntary labour mobilisation for megalith building." Quoted from the accepted manuscript.]
16. Madgwick, R., Lamb, A. L., Sloane, H., Nederbragt, A. J., Albarella, U., Parker Pearson, M., & Evans, J. A. (2019). Multi-isotope analysis reveals that feasts in the Stonehenge environs and across Wessex drew people and animals from throughout Britain. Science Advances, 5(3), eaau6078. doi:10.1126/sciadv.aau6078 [131 pigs from four Late Neolithic complexes; "few pigs [were] raised locally"; five are "likely to derive from Scotland."]
17. Dietrich, O., Heun, M., Notroff, J., Schmidt, K., & Zarnkow, M. (2012). The role of cult and feasting in the emergence of Neolithic communities. New evidence from Göbekli Tepe, south-eastern Turkey. Antiquity, 86(333), 674–695. doi:10.1017/S0003598X00047840 [pp. 684–687: "there had to be a means of bringing together groups from different areas and organising communal work," and the answer "lies in the widespread evidence for extensive feasting."]
18. Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J., et al. (2019). Cereal processing at Early Neolithic Göbekli Tepe, southeastern Turkey. PLoS ONE, 14(5), e0215214. doi:10.1371/journal.pone.0215214 [dates in the 10th and 9th millennia BC; "a smaller group of people actually lived at Göbekli Tepe permanently." Banning, E. B. (2011), Current Anthropology, 52(5), 619–660, first argued that its buildings may be houses.]
19. Schelling, T. C. (1960). The Strategy of Conflict. Harvard University Press, ch. 3, p. 57. ["some focal point for each person's expectation of what the other expects him to expect to be expected to do"]
20. Chwe, M. S.-Y. (2001). Rational Ritual: Culture, Coordination, and Common Knowledge. Princeton University Press, pp. 3–4. ["Public rituals, rallies, and ceremonies generate the necessary common knowledge."]
21. Šprajc, I., Inomata, T., & Aveni, A. F. (2023). Origins of Mesoamerican astronomy and calendar: Evidence from the Olmec and Maya regions. Science Advances, 9(1), eabq7675. doi:10.1126/sciadv.abq7675 [Discussion: "To coordinate ritual schedules among participants from ever broader areas, the builders of ceremonial complexes needed to elaborate solar observational calendars." The complexes date from 1100 to 750 BCE.]
22. Dietler, M., & Herbich, I. (2001). Feasts and labor mobilization: Dissecting a fundamental economic practice. In M. Dietler & B. Hayden (Eds.), Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power (pp. 240–264). Smithsonian Institution Press. [p. 256: work feasts are "an exchange of labor for hospitality."]
23. Donald, M. (1993). Précis of Origins of the Modern Mind: Three stages in the evolution of culture and cognition. Behavioral and Brain Sciences, 16(4), 737–748. doi:10.1017/S0140525X00032647 [external symbolic storage systems "allow humans to circumvent, at least partially, the limitations of biological working memory"; the third transition began "in the late Upper Paleolithic with the invention of the first permanent visual symbols." Quoted from the archived preprint.]
24. Englund, R. K. (1988). Administrative timekeeping in ancient Mesopotamia. Journal of the Economic and Social History of the Orient, 31(2), 121–185. doi:10.1163/156852088X00070 [p. 122: "a twelve-month, 360-day year in the archaic documents from the end of the 4th millennium"; p. 163: the bowl sign NINDA for "one day's grain ration"; p. 164: the month as "the unit basic to rationing systems"; p. 125: Ur III workday ledgers (about 2100 BC) count in 30-day months.]