467
Documenta Praehistorica XLIII (2016)
What is a lunar standstill III 1|
Lionel Sims
University of East London, London, UK
lionel.sims@btinternet.com
Introduction
Thom’s publications between 1954 and 1975 galva-
nised a generation – with horror among many ar-
chaeologists and inspiration for others, some of
whom began reconstructing the discipline of ar-
chaeoastronomy (Ruggles 1999.1–11, 275). How-
ever, since the 1980s, an extended re-evaluation of
his work within archaeoastronomy has questioned
many of his claims that some monuments of prehi-
storic North-West Europe were aligned on the Sun’s
solstices and the Moon’s standstills (Ruggles 1999.
49–67). Nevertheless, some leading archaeologists
in Britain have begun to embrace the evidence for
prehistoric monuments’ ‘astronomical’ alignments
(Thomas 1999; Parker Pearson 2012). To unravel
what has been lost and gained after half a century
of scholarly endeavour, we need to re-examine the
properties of prehistoric monument ‘astronomical’
alignments to trace their variable engagement by
different researchers. 
Solar and lunar alignments
A prehistoric monument alignment on the Sun and
Moon is an architectural arrangement of materials
orientated on a horizon position locating where
either luminary rises or sets. At summer solstice,
the Sun rises in the north-east and sets in the north-
west. Six months later at winter solstice, the Sun
rises in the south-east and sets in the south-west. For
any given time period, latitude and horizon altitude
these horizon positions are fixed. At the latitude of
Stonehenge, the azimuths of the solstice Sun are
ABSTRACT – Prehistoric monument alignments on lunar standstills are currently understood for hori-
zon range, perturbation event, crossover event, eclipse prediction, solstice full Moon and the solari-
sation of the dark Moon. The first five models are found to fail the criteria of archaeoastronomy field
methods. The final model of lunar-solar conflation draws upon all the observed components of lunar
standstills – solarised reverse phased sidereal Moons culminating in solstice dark Moons in a roughly
nine-year alternating cycle between major and minor standstills. This lunar-solar conflation model
is a syncretic overlay upon an antecedent Palaeolithic template for lunar scheduled rituals and
amenable to transformation. 
IZVLE∞EK – Poravnava prazgodovinskih spomenikov na Lunine zastoje se trenutno sklepa za razpon
horizonta, perturbacijo, prehod Lune, napovedovanje Luninih mrkov, polne lune na solsticij in sola-
rizacije mlaja. Prvih pet modelov ne zadostuje kriterijem arheoastronomskih terenskih metod. Zad-
nji model lunarno-solarne zamenjave temelji na vseh opazovanih komponentah Luninega zastoja –
solarizirane obrnjene Lune, ki kulminirajo v mlaju v solsticiju v devetletnem izmeni≠nem ciklu med
najve≠jim in najmanj∏im zastojem. Ta lunarno-solarni model je postavljen na temeljih predhodne
paleolitske predloge za na≠rtovanje lunarnih ritualov in je dojemljiv za preobrazbe.
KEY WORDS – lunar standstill; synodic; sidereal; lunar-solar; syncretic
1 The first version was my 2007 paper for the 2006 SEAC conference in Rhodes, and the second was an unpublished paper for the
2014 SEAC conference in Malta. This latest, third, paper is an elaborated version of the second paper. It should be read alongside
the Forum debate ‘What is the minor standstill of the Moon?’ in Skyscape Archaeology 2(1): 67–102.
DOI> 10.4312\dp.43.24 
Lionel Sims
468
about 40° above and below the west-east axis (Fig.
1). From three days before and after the solstice, the
unaided human eye cannot detect the 2´ change
from its solstice limit, and in seven days it remains
within 11´ of its solstice extreme (Ruggles 1999.24–
25). Each of the four solstice rise and set events
therefore lasts for at least one week in prehistory.
Outside of the two weeks of both solstices, for about
51/2 months, the Sun’s horizon rise and set positions
slowly begin to change towards its greatest rate of
daily horizon change during the equinoxes of 24´
per day – three quarters of its 32´ diameter. A mo-
nument’s single solstice alignment therefore abstracts
out just 1 horizon event repeating for seven days
out of a possible 1782 solar alignments. While a se-
vere selection of the Sun’s movements, nevertheless
these horizon positions do not contradict, and may
act as proxies, for the seasonal changes marked by
the full suite of the Sun’s movements.
A monumental lunar alignment is a more complex
matter and, in important respects, quite different
from a solar alignment. For while the Moon also has
horizon range limits just as the Sun does, instead of
being fixed they vary over an 18.61 year cycle. Thom
named this phenomenon the lunar standstills, which
every 9.3 years alternate between major and minor
standstills (Thom 1971.19, 124, 173). The range of
the Moon’s horizon rises and sets complete over the
course of the 27.3 day sidereal month – the time it
takes to complete its circulation of the earth. During
a major standstill of the Moon at the latitude of Sto-
nehenge, the Moon rise and sets about 10° of azimuth
further north and south on the horizon than the Sun
ever reaches during its solstices. After a period of
over a year, the Moon’s orbital plane
about the earth begins to reduce
from 5° above the earth’s equatorial
plane about the Sun to, 9.3 years
later, 5° beneath the equatorial
plane. This minor standstill’s hori-
zon range limits are now about 10°
within the Sun’s solstice positions
at the latitude of Stonehenge, and
for over a year the Moon returns to
these ‘same’ positions (Fig. 1). The
Moon therefore has eight horizon
limits, unlike the Sun’s four. 
Throughout its 18.61 year cycle, the
Moon rises and sets on intermedi-
ate horizon positions between these changing hori-
zon extremes of the Moon. Only during the period
of a lunar standstill, major or minor, does the Moon
return to the same small region of the horizon, when
it is at its standstill range limit. It is therefore a selec-
tion of just one horizon moment during one day or
night out of about 27 days and nights during two
periods of the 18.61 year cycle. While, like the Sun,
a lunar alignment is therefore a selection from all
possible horizon positions, unlike the Sun, it is not
an observation repeated over the course of a week,
but a one-moment time-lapsed observation separat-
ed by periods of about 27 days during either the ma-
jor or the minor standstill about every nine years.
When the Moon returns each sidereal month to its
horizon alignment, it does not repeat its earlier lunar
phase. While it takes the Moon 27.3 days to circle
the earth, during that time the earth is also moving
around the sun, and the three bodies re-align at the
completion of the synodic month in 29.5 days. There-
fore, a lunar alignment presents phase sequences
which are about 2.2 days earlier phases than the
previous lunar phase alignments. An alignment on
a lunar standstill therefore selects lunar phases in
reverse sequence to when, contrarily, we view the
Moon’s phases over the course of a synodic month.
Lastly, as the Moon circulates the Earth, which itself
circulates the Sun, the changing gravitational inter-
action of all three bodies are reflected in small dif-
ferences in the position of the maximum range lim-
its of either standstill. Modern heliocentric position-
al astronomy has shown that at different positions
in its orbit of Earth, these perturbations exhibit a
regular sinusoidal alternation of the order of about
10´ of arc (Thom 1971.15–27; Sims 2006). 
2 Subtracting the 14 days of the two solstice weeks from 365 days (=351), dividing by two for the outgoing and returning Sun over-
lapping on the same alignment (=175.5), and then adding 2 for the two solstice events, each of one week (=178).
Fig. 1. Plan view of natural horizon alignments on Sun’s solstices
and Moon’s standstills at Stonehenge 2500BC (bracketed numbers
are degrees of azimuth above or below the west-east axis).
What is a lunar standstill III|
469
In summary, just like a solstice alignment, a lunar
standstill alignment is a selection of a few horizon
positions from its total rise or set positions, but un-
like the Sun, this alignment is not a simple proxy for
other characteristics of the Moon. Instead of four sol-
stice horizon positions, there are eight lunar stand-
still horizon range limits of the major and minor
standstills; its sidereal properties are cryptic, not re-
dundantly obvious as are the solstices; they reverse
its lunar phases, and at its horizon limits exhibit
small perturbations. It is in the variable interpreta-
tion of these similarities and differences between
solstices and standstills that scholars in archaeoas-
tronomy have marked out six different approaches
to lunar standstills. There is therefore no consensus
within archaeoastronomy on how to interpret lunar
standstills, and these divisions require resolution
(Sims 2013a).
To evaluate the strengths and weaknesses of each of
these six positions, I will refer to the archaeological
details at the Stonehenge monument complex, since
this is, arguably, one of the culminating achievements
of a wider cosmology extant throughout Neolithic
Europe. This late Neolithic and Early Bronze Age site
has attracted close scholarly attention, which has
amassed enough fine detail to allow us to design dis-
criminating tests between the available interpretive
models.
Lunar standstills as horizon range
According to Fabio Silva and Fernando Pimenta (2012.
206), the majority view within archaeoastronomy is
that lunar standstills were selected by prehistoric
monument builders for the Moon’s extended hori-
zon range displayed only during the major stand-
still of the moon. Throughout this period of over a
year, every thirteen or so days the Moon reaches its
northern and southern limits at azimuths roughly
50° above and below the west-east axis at the lati-
tude of Stonehenge, which is about 10° beyond the
Sun at its solstice horizon positions (Fig. 1). Between
these periods, the Moon’s daily and nightly changes
touch intermediate horizon positions and pass
through the range limits of the minor standstill,
which the Moon reaches 9.3 years later. When the
Moon is at its minor standstill, its azimuth is about
30° above and below the west-east line, well within
the Sun’s azimuth at about 40°. Compared to the
major standstill, this approach considers that since
the Moon passes through this position for much of
the 18.61 year cycle, the minor standstill is both dif-
ficult to observe and has little significance because
of the narrower width of its horizon range. Gauging
the lunar standstills against the Sun’s horizon range
is therefore a solarist attribution that the Moon at a
major standstill is a ‘superior sun’ by virtue of wide
horizon range alone. Authors who adopt this view
emphasise the period of 18.61 years as the time span
of lunar standstills.
However, César González-García (2016) has shown
that it is not difficult to observe the minor standstill
of the Moon. During this period, the Moon’s horizon
range is at its narrowest, while the number or rises
and sets remain the same at just over 27 events each
sidereal month. This produces a proportionally in-
creased bunching of maximum horizon events which
is greater than the maximum horizon range bunch-
ing for both the Moon’s major standstill and the
Sun’s solstices. Further, if the solarist understand-
ing of the prehistoric appropriation of lunar stand-
stills is correct, then it would predict that no monu-
ments in prehistory would have alignments on the
minor standstill of the Moon beyond those of chance.
This is incorrect. Clive Ruggles (1999.74–76, 95) has
shown that Scottish recumbent and West Scotland
rows have statistically significant alignments on both
the southern major and minor standstills, as has
Aubrey Burl (1981) for the Clava Cairns. Further-
more, along the West Kennet Avenue at Avebury,
Lionel Sims (2015) has shown that across paired
stones 1–37, out of 362 possible alignments, 145
solar and lunar alignments can be found. Using Ber-
noulli’s Law, so accounting for possible overstate-
ment from straight sections along the avenue with
a regular horizon, Sims shows that this is a possible
chance occurrence for 3.3 in every million times.
Within these 145 alignments, 21 are on solstices, 31
on major standstills and 48 are on minor standstills.
The rejection of minor standstills as chance align-
ments therefore does not meet the requirements ex-
pected in archaeoastronomy field methods. Empha-
sising the Moon’s extended horizon range during
the major standstill is more an extemporary field
method device to distinguish a monument’s possi-
ble orientation on the Sun. But in its rejection of the
minor standstill of the Moon, this approach has failed
to lay the basis for investigating what properties it
may share with the major standstill.
Major and minor standstills as separate pheno-
menon
The second type of archaeoastronomy theory con-
siders the minor standstill of the Moon as requiring
special explanation separate from that of the major
Lionel Sims
470
standstill, which is understood to be defined by its
wide horizon range. There are currently two versions
of this approach – North’s (1996) alternating pertur-
bations and Silva’s and Pimenta’s (2012) crossover
model.
North showed that, while in plan view Stonehenge
is gaps surrounding an empty space, in elevation
view, paradoxically, it is an ‘obscuration’ device of a
seemingly near-solid mass of stone when approa-
ching it from the north-east uphill along the Ave-
nue and past the Heel Stone. Viewed from the Heel
Stone, two windows can be seen one above the
other, framed by the Grand Trilithon and the outer-
circle lintels (Fig. 2). From the right-hand side of the
Heel Stone, the lower window frames the setting
winter solstice sunsets, and from the left-hand side
of the Heel Stone at an altitude of about 4–5°, the up-
per window captures the setting southern minor
standstill moonsets. North considered these two
alignments separately and suggested that the south-
ern minor standstill moonset perturbations would
have been seen in this upper window smoothly zig-
zagging left and right four or so times over the course
of a minor standstill year. North suggested that since
the Sun does not do this when at its solstice horizon
range limit, then the people of Stonehenge would
have considered this property of the southern minor
standstill magical and worthy of a monument (North
1996.441–474; Sims 2006).
It is not clear that the Moon’s perturbations will be
observed at Stonehenge, as the archi-
tecture of the monument does not
assist such high-fidelity observations.
The distance from the Heel Stone to
the Grand Trilithon is about 88m,
which is not enough to provide the
resolution to discern the movements
of a few minutes of arc of any (ob-
scured) edge of the lunar disc. From
the Heel Stone, the height of the up-
per window would have subtended
an angle of about 10´ and a width
over one degree. Since the lunar disc
is about 31´ in diameter, this upper
window would have framed just a
descending sliver of the Moon in its
1–2° of oscillating azimuths at its
standstill horizon range limits (see
below). That is assuming the perturbations could be
seen distinct from other movements of the Moon.
As the Moon is constantly changing its declination3,
and since the moment of the geocentric declination
extreme is generally independent of the moment
when the Moon meets the horizon, by the time the
Moon is on any horizon, its declination is at a dif-
ferent value from its perturbation limit. Therefore,
there will be no regular zigzag horizon movement of
the Moon at its minor standstill limit, but irregular
movements. This can be seen in Figure 3, which con-
trasts the smooth sinusoidal perturbations of the
Moon when orbiting the Earth with how these be-
come irregular oscillations by the time the Moon sets
on the horizon (see also Sims 2007).
Finally, every standstill at its horizon limit, whether
major or minor, north or south, rising or setting, ex-
hibits similar (irregular) perturbations to that of the
southern minor standstill moonsets, so either all of
these are equally ‘magical’ or none of them are. North
defines the southern minor standstill by the bound-
ary oscillations of the lunar horizon range limits, and
the major standstill by its greater horizon range than
that of the Sun, without interrogating the shared
properties of both. This model also therefore empha-
sises the period of 18.61 years between major stand-
stills, and misinterprets minor standstills.
Silva’s and Pimenta’s winter solstice lunar crescent
crossover model, the second within this group, offers
an alternative explanation to that of alignments on
3 This is the measure used in modern heliocentric positional astronomy, and is the number of degrees above or below the celestial
equator. In the geocentric flat earth cosmology of prehistory, a concept of declination would have carried no meaning. Instead,
some concepts equivalent to our ‘azimuth’ and ‘altitude’ would have been used.
Fig. 2. The lower and upper gaps through the central axial stones
at Stonehenge as seen from the left- and right-hand sides of the
Heel Stone (North 1996.446). Numbering follows the accepted no-
tation for each stone.
What is a lunar standstill III|
471
the minor standstill of the Moon. They demonstrate
that during summer and autumn, the first crescent
Moon always sets to the left of the Sun’s horizon
setting, but within a period of 150 days over winter
the Moon switches and sets to the right of the Sun’s
setting position. The pattern of alignments for this
crossover event is a non-Guassian distribution with
a marked modal frequency which coincides with that
of the southern minor standstill of the moon. Any
monument with an alignment on the minor stand-
still would, the authors suggest, be better explained
by this annual event. According to the authors, the
preferred function for the crossover event would
have been to mark annual calendrical calculations,
while the major standstill of the Moon is best explain-
ed by its wide horizon range (Silva, Pimenta 2012.
202, 206).
While the authors claim that the Sun crosses over
the first crescent Moon at the winter solstice, dur-
ing the 2014/2015 minor standstill of the Moon the
actual date of the crossover was the 12th December,
not the solstice of 21st December. Not just the time,
but the horizon position of the crossover event is
also highly variable. While the azimuth band for
the range limit of the minor standstill is only about
1–2° (see below), that of the first crescent winter
crossover horizon event is spread over an azimuth
range of about 26°, a qualitatively higher order of
magnitude than that of all lunar standstills. The Sto-
nehenge axial upper window, which captures all the
southern minor standstill moonsets, would not be
able to capture about 89% of these crossovers.
It remains to be explained why the non-Gaussian
distribution of the first lunar crescent crossover with
the Sun has a peak which coincides with the south-
ern minor standstill horizon moon-
sets. A crossover event is an amal-
gam of the properties of two vari-
ables – the movements of the Sun
and those of the Moon. To disentan-
gle conjoint influences, we need to
elaborate their separate effects. Con-
sider first when the Moon is at the
southern minor standstill. For this
year and adjacent years, the Moon’s
horizon range does not move much
further south than when within 10°
of azimuth or so of the Sun’s horizon
range. Only when the Sun starts to
leave its winter solstice declination
and begins to set further north on
the horizon does it approach close
enough to the Moon for a crossover to occur.
Therefore, while the event will coincide or be close
to the southern minor standstill declination, it can-
not take place at the winter solstice. Since Stone-
henge has a lower window aligned on winter sol-
stice sunset, the crossover model cannot explain the
monument’s conflation of the Moon and Sun. Now
consider when the Moon is at the major standstill.
The Moon then sets much further south on the ho-
rizon than the Sun ever reaches, but the Moon’s ho-
rizon movements from south to north take 13/14
days, while those of the Sun take 6 months. The
Moon’s movements will now allow the crossover
event to take place when the Sun is still at, or close
to, its winter solstice horizon position, but not at the
horizon position of the southern minor standstill.
Therefore, this event will coincide with a winter sol-
stice declination, but it cannot take place when the
Moon is at the southern minor standstill declination.
We now turn to the Moon’s movements when either
expanding or contracting its horizon range and can
cross over with the Sun. During this inter-standstill
period, when the Sun has moved to the horizon posi-
tion of the southern minor standstill about 6 weeks
after the winter solstice, the Moon will cross over
the Sun once it has left its southern horizon limit. In
both cases, the Sun is not the winter solstice Sun,
and neither is the Moon the southern minor moon.
In aggregate, these separate effects create the non-
Gaussian distribution, with a peak that coincides
with the southern minor standstill, yet it will only
rarely take place at winter solstice; most frequently,
it will take place when the Moon is not at the south-
ern minor standstill, and it will never take place dur-
ing the southern minor standstill at winter solstice.
Fig. 3. Contrast between the Geocentric Extreme Declination of the
Moon (GED) 2491–2488 BC and Azimuth (Az) of the southern mi-
nor standstill Moon setting in the Upper Window at Stonehenge. 
Note: Azimuth values have been transformed to fit the declination
scale, and should be examined for shape, not values, alone.
Lionel Sims
472
Therefore, a monument aligned on the minor stand-
still of the Moon and the winter solstice Sun cannot
be explained by the crossover model.
The shared properties of Major and Minor lu-
nar standstills
The third group of theories for lunar standstills iden-
tifies shared properties between the major and mi-
nor standstills; three different theories take this ap-
proach.
Alexander Thom (1971.15) suggested that monu-
ments aligned on both types of standstill were de-
vices to predict lunar eclipses. In both cases, eclips-
es occur when the Moon’s perturbations coincide
and add to the horizon range limits of lunar stand-
stills every 9.3 years. Since the maximum perturba-
tion only very rarely occurs when the Moon is on
the horizon, Thom suggested that ‘extrapolation de-
vices’ were necessary to extrapolate the precise mo-
ment of the maximum. There is, however, no evi-
dence for such devices, and it would not be possible
to track these tiny horizon movements with low-
resolution alignments (Ruggles 1999.63). Neverthe-
less, as during a lunar standstill, these maximum
perturbations always occur close to the equinoxes,
and as NW European late Neolithic/EBA prehistoric
monuments which conflate the Moon with the Sun
choose the solstice Sun, and not its equinoxes (Rug-
gles 1999.148–151), the monument builders appear
to be avoiding not predicting an eclipsed moon. Since
an eclipsed Moon is an interrupted full Moon, it seems
that the monument builders were selecting for some
aspect of an uninterrupted cycle of lunar phases at
their ritual centres.
The remaining two models of lunar standstills cur-
rent within archaeoastronomy engage with the in-
trinsic properties of the Moon’s phases rather than
just their horizon location. Ruggles suggests that the
monument builders entrained their monuments on
lunar standstill full Moons. Ruggles’ critique of Thom’s
work was based on a major field work exercise of
five regional groups of monuments in the British
Isles. A strong preference emerged from the data for
lunar alignments on the southern lunar standstills,
both major and minor, and particularly onto settings
rather than risings. Ruggles concluded that they
would have observed the full Moon setting along
these alignments (Ruggles 1999.75, 96–98, 107,
128, 130, 138–9, 149, 154). But for the Moon to be
full when it is at its southern extreme range, whether
at the major or the minor standstill, the Sun has to
be at its northern extreme, namely at summer sol-
stice. Since all of these monument groups also includ-
ed alignments on the winter solstice sunsets, not
summer solstice, and since they paired the southern
standstill moons with the winter solstice, this sky-
scape archaeology provides no justification for con-
flating the southern standstill Moon as full Moon
with the summer solstice. Furthermore, the Sun
would not be simultaneously setting in the lower
window, but rising on the north-east horizon. Both
the astronomy of a single point estimate for a lunar
standstill and the skyscape archaeology of Stone-
henge do not support selecting summer solstice full
Moon as the way to interpret the southern minor
standstill at Stonehenge (see also Heggie 1981.98).
The final sixth model of lunar standstills addresses
the emergent properties from their combination with
the Sun’s solstices. Sarsen Stonehenge is a binary
monument, with two circles and two horseshoes.
With the half height, half width and half depth stone
11 it has 291/2 stones in the outer sarsen circle and
19 stones in the bluestone horseshoe, both lunar
numbers. The building’s design principle of obscu-
ration by nested lintelled and ranked wide pillars
allows bringing onto one alignment from the Heel
Stone a lower window opening onto the setting win-
ter solstice Sun and above it in another window up-
on the southern minor standstill moonsets (North
1996; Sims 2006; 2016). In short the precise archa-
eology of sarsen Stonehenge defines what ‘astron-
omy’ is being selected for through the cropping of
the lunar and solar discs by these two windows.
Both Ruggles (1999) and John North (1996) define
a lunar standstill by a single point for its geocentric
extreme, and respectively use this value to justify
their suggestions that the builders either selected
full Moon at summer solstice or alternating pertur-
bations during the minor standstill. To evaluate the
details of these positions, consider in Figure 4 the
southern minor standstill of the Moon at the latitude
of Stonehenge for the thirty months from 9th Novem-
ber4 2510 to 30th April 2507 BC – the minor stand-
still immediately before the probable date for the
building of sarsen Stonehenge around 2500 BC (Par-
ker-Pearson 2012).
The extreme range is reached on 12th April 2508
BC at a declination of –19.25°, when the Moon is a
4 All dates are Julian dates.
What is a lunar standstill III|
473
waning quarter Moon. Full Moon is met three side-
real months later, during the summer solstice week
of the 3rd July5 2508 BC at a declination of –19.59°.
Since one degree of declination translates into about
two degrees of azimuth at this latitude, ignoring any
refraction effects, this summer solstice full Moon is
about 41´ of azimuth less than the extreme point of
the standstill. A descending sliver of this Moon would
still be seen in the Stonehenge upper window with
its declination limits as shown in Figure 4, but to
the left by over one lunar diameter and not at the
actual extreme value used to define the lunar stand-
still. This value cannot justify the choice of either full
Moon or alternating perturbations as suggested by
these two authors. Furthermore, during full Moon,
the Sun would not be simultaneously setting in the
lower window, but rising on the north-east horizon.
Both the astronomy of a single point estimate for a
lunar standstill and the skyscape archaeology of Sto-
nehenge do not support selecting summer solstice
full Moon as the way to interpret the southern minor
standstill at Stonehenge (see also Heggie 1981.98).
Over a period of about 36 months, the southern ex-
treme horizon moonsets of the minor standstill of
the Moon irregularly alternate between a declination
of about –19° and –20° (Fig. 4). Unlike the sun’s sol-
stice horizon position, a lunar standstill extreme limit
is not a point, but a small region on the horizon
which displays a vacillating reverse sequence of 36
or so lunar phases. These phases of the Moon include
all phases that can be witnessed during the transit-
ing synodic Moon, but in reverse. Selecting any one
of these phases to define a lunar standstill cannot be
justified by a single declination value. The archae-
ology at Stonehenge, and at the five regional sites
reported by Ruggles, is clear – all of these 36 or so
reversed lunar phases descend through the upper
window at Stonehenge. This includes the dark Moon
of –20.115° during the week of winter solstice on
10th January 2507 BC, ten sidereal months after the
extreme declination of the Moon (Fig. 4). But the
monument builders conflated this sequence in the
upper window with the week of winter solstice sun-
sets in the lower window. By the binary architecture
of the monument, this arrangement selects the entire
reverse phased sequence of lunations which culmi-
nate with the binary astronomy of dark Moon at
winter solstice. Since the subsequent Moons on this
alignment are reversed waning crescent Moons, and
as these are visible only when rising a few hours be-
fore sunrise and not at their setting, then for three
or so months from the winter solstice no Moon will
be seen setting in the Stonehenge upper window. It
seems that the monument builders required their
rituals to coincide with the start of the longest dark-
est night.
It may be thought that when a dark Moon coincides
with the Sun’s solstices outside the periods of lunar
standstills, this invalidates selecting this conjunction
for standstills alone. For example, dark Moons occur-
red during this millennium’s winter solstice week in
2003, 2011 and will do so in 2017, none of which
are standstill years. However, the issue is that none
of these occurrences outside of the 2014–2016 pe-
riod of the lunar standstill could be identified by a
stable alignment spread of a few degrees over the
course of 36 months. The moon’s horizon range lim-
its occupy two very small and stable parts of the
horizon only at these periods of the major and mi-
nor standstills. This is not possible during inter-stan-
dstill years, where the range limits are changing sub-
stantially. In Figure 5, it can be seen that over the
three years of the inter-standstill years of 2505–
2503 BC the southern lunar extremes range from
–22.4° to –27.5° declination. This difference of 5.1°
declination equals 10.2° of horizon azimuth, and
could not be accommodated within the Stonehenge
architecture. Only during the period of a lunar stand-
still could an alignment with a range of two degrees
Fig. 4. Southern Minor Standstill 2510–2507 BC at
latitude of Stonehenge. Note: 1 Vertical axis in de-
grees of topocentric declination. 2 Horizontal axis
is the number of each southern lunar extreme
between 2510 and 2507 BC. The topocentric ex-
treme declination occurred on 12th April 2508 BC
at declination –19.25°. 3 Dark Moon during week
of winter solstice (in red) on 10th January 2507 BC
at declination –20.115°. 4 Declination values for
all moonsets at 4 p.m., the approximate time the
winter solstice Sun sets in the Stonehenge lower
window. 5 Source SkyMap Vers. 8.
5 Summer and winter solstices were on 3rd July and 9th January, respectively, around 2600 BC, and 2nd July and 8th January, res-
pectively, around 2500 BC.
Lionel Sims
474
accommodate a horizon reversed suite of lunar
phases culminating in dark Moon at winter solstice
and so be enshrined into the axial centre of the mo-
numents. Such was the case for the minor standstill
period of 2014–2016, when dark Moons also occur-
red during winter solstice week on the 22nd Decem-
ber 2014, when the Sun was setting in the Stone-
henge lower window.
The peaks and troughs in the Moon’s horizon mo-
vements will therefore migrate serpent-like through
each standstill, and this would seem to invalidate
the model’s claim that dark moons will always coin-
cide with solstice weeks during lunar standstills.
This misunderstanding of the model is a conse-
quence of defining lunar standstills by the 18.61
year cycle of modern celestial mechanics, whereas
prehistoric sky watchers understood it by horizon
azimuth. However it should be remembered that
the archaeology of Stonehenge, and the five
regional groups studied by Ruggles, specify the
combined selection of winter solstice sunsets and
the southern (major and minor – see below) lunar
standstill sunsets. The association of these two
alignments migrates through a 36 month period of
each standstill, sometimes occurring before the
extreme declination and at other standstills after
that extreme. Table 1 shows the six minor stand-
stills that occurred during the century before the
presently understood building date of Stonehenge.
At the latitude of Stonehenge, in these cases over
a maximum period of twenty months, the moment
of the extreme declination of the Moon varies over
the course of the one month of April, while dark
Moon coinciding with winter solstice also varies
over the course of the third millennium BC, but by
either four months before the extreme or ten
months after it. In short, within a small alignment
horizon band of about 11/2° (0.7196 x 2) over the
course of one year in the three years of these mi-
nor standstills, 13 reversed phased sidereal Moons
culminate in dark Moon at winter solstice for all six
minor standstills.
Northern lunar standstills
Our critique through the details of sarsen Stone-
henge of the first five archaeoastronomy models of
lunar standstills has concentrated on the minor
standstill of the Moon at its southern moonsets. An
identical structure is revealed in the northern ex-
treme moonsets of the minor standstill and in the
southern and northern major lunar standstills. Fi-
gure 6 shows the northern and southern minor
standstill for 1978 AD and similarly for the major
Fig. 5. Southern lunar extremes at the latitude of
Stonehenge during the inter-standstill years 2505–
2503 BC. (Source SkyMap vers. 8). Note: 1 Vertical
axis in degrees of topocentric declination. 2 Ho-
rizontal axis is the number of each southern lunar
extreme between 2505 and 2503 BC. 3 Declina-
tion values for all moonsets at 4 p.m., the approx-
imate time the winter solstice Sun sets in the Stone-
henge lower window. 4 Source SkyMap Vers. 8. 5
Another possible challenge concerns the claim for
the meshing of lunar phases exactly conflating
dark Moon with the solstices between each stand-
still. Since the standstill cycle recurs every 18.61
years, and as this is an irregular number, repeat-
ing standstill dark Moons will not always coincide
with the solstices close to the central extreme decli-
nation value of the following or preceding stand-
still. 
Year BC Moon’s Moon’s Winter solstice Winter solstice Deviation in Deviation of
topocentric topocentric dark Moon dark Moon sidereal months winter dark Moon from
extreme extreme date declination° of dark Moon topocentric extreme
date declination° from top. Extreme in degrees declination°
2602 30 April –19.633 10 January –20.0101 –4 0.3768
2583 2 April –19.517 10 January –19.808 –3 0.3713
2564 28 April –19.241 9 January –19.727 –4 0.4853
2546 11 April –19.367 10 Jan 2545 –19.819 +10 0.4519
2527 11 April –19.233 10 Jan 2526 –19.953 +10 0.7196
2508 12 April –19.583 10 Jan 2507 –20.116 +10 0.5325
Tab. 1. Relationship between dark Moon during week of winter solstice and the topocentric extreme of
the southern minor standstills at latitude of Stonehenge for 2602–2508 BC. (Source SkyMap vers. 8)
What is a lunar standstill III|
475
standstill of 1969 AD. It should be re-
membered that these smooth sinusoidal
plots represent the points of the geo-
centric extremes of the Moon during its
orbit around the Earth, but that once
reaching the horizon, the Moon’s decli-
nation will have changed and a more
erratic fluctuation will be observed (see
Figs. 3 and 4). The remaining pattern
and properties shown in Figure 6 are ge-
neral for all lunar standstills (see Thom
1971.20; 1978.11–16; North 1996.553–
572).
Just as we found that the extreme dec-
lination value for the southern minor
standstills occurred at first or third quar-
ter Moons around the time of the equi-
noxes, this remains the case for north-
ern minor standstills and major stand-
stills at both northern and southern ex-
tremes. For the twenty or so sidereal lu-
nations shown here in each lunar stand-
still, they all occur within a very narrow
band of about 30´ of declination and all
present a reverse sequence of lunar pha-
ses every 27 or so days spread over the
course of 11/2 years. The remaining
structural feature of dark Moon syn-
chrony with the Sun’s solstice that we
have found for the southern minor lu-
nar standstills also belongs to the north-
ern minor standstills and both the south-
ern and northern major lunar standstills.
Thus while during the southern major
standstill dark Moon happens during the
week of winter solstice just as it did for
the southern minor standstill, now for
the northern lunar standstills, whether
minor or major, dark Moon occurs dur-
ing the week of summer solstice.
For the minor standstill period of 2014–2016 dark
Moons occurred not just during winter solstice week
on the 22nd December 2014, but also for the sum-
mer solstice week of 17th June 2015. This synchro-
nisation will always occur within a period of about
36 months around the extreme horizon moonsets of
each lunar standstill of the Moon. Because of this
6 Four considerations need to be kept in mind when interpreting this computer- generated model of lunar standstills: (a) The verti-
cal axis is cropped, so abutting the northern and southern horizon extremes when they are actually widely separated on the west-
ern and eastern horizons by about 60° for minor standstills and about 100° for major standstills at the latitude of Stonehenge.
(b) The diameter of the lunar orb is not to scale and has been reduced to 5’ from its actual 30’ to better display its 20’ perturba-
tions. (c) The Moon’s path is measured by its geocentric declination, which is a modern measure from heliocentric positional astron-
omy of the distance in degrees from the celestial equator to the centre of the lunar disc. Prehistoric horizon flat-earth ‘astronomy’
measures by azimuth and horizon altitude to the upper or lower limb of the Sun and the Moon. At the latitude of Stonehenge, 1°
of declination translates as about 2° of azimuth for a given horizon. (d) The Moon’s geocentric extreme declination occurs at any
time in its transit around the earth and only very rarely when it is breaking the horizon. As the Moon is always changing its dec-
lination, the regular sinusoidal variation in perturbation displayed in this figure is not observable on the horizon.
Fig. 6. Monthly geocentric extreme declinations of the 1969
major standstill and the 1978 minor standstill, by date and
lunar phase (Morrison 1980)6.
Lionel Sims
476
fluctuating synchronisation of solstice and dark
Moon during each minor and major standstill, the
alternation between them is about nine years and
the 18.61 year nodal cycle length is an understand-
ing from modern heliocentric positional astronomy
and not relevant to a prehistoric flat-earth geocentric
cosmology.
Interpreting the model of dark Moon lunar-
solar conflation
This understanding of lunar standstills as conflating
the Sun’s solstices with dark Moon seems counter-
intuitive for observational astronomy – “... this
makes no sense...[when] the Moon is new, and
hence invisible ...”(Ruggles 1999.247). It was pro-
bably for this reason that Ruggles characterised his
own field data, which largely conflate solstice and
standstill alignments to the south west, as ‘anom-
alous’ (Ruggles 1999.142, 158). But as long as his
findings are data, then it is only the interpretation
of full Moon that can be anomalous. Instead of mak-
ing an a priori assumption that the monument
builders required a full Moon, we need to ask the
question of what properties are revealed by solaris-
ing the Moon’ alignment? Sims’ model of lunar-solar
conflation argues that a monument’s combined align-
ment on the Moon and the solstice Sun selects side-
real properties that reverses lunar phases, attenu-
ates a full suite of lunar phases in a time-lapsed ob-
servation exercise spread over the course of a year,
and culminates at winter and summer solstices with
dark Moon at southern and northern lunar stand-
stills, respectively, every nine or so years. In short,
this model embraces all of the observed characteris-
tics of lunar standstills as a horizon alignment. We
would expect these themes of solarisation, reversal,
attenuation and alternating dark Moons over a nine-
yearly cycle to inform an interpretive theory of lunar
standstills.
The suite of properties which has survived our cri-
tique exponentially reduces the number of possible
interpretations. Any theoretical model which priori-
tises full Moon, annual calendars or eclipse predic-
tion is not supported by our findings. This complex
portfolio of characteristics suggests symbolic aspects
of ritual embedded within signalling the large la-
bour investments required by monumental architec-
ture. If this interpretation is correct, then we would
expect similar themes of alternation, reversal and
calibration or re-calibration of the lunar and solar
cycles to be predicted by an appropriate theoretical
model.
Three considerations suggest we should look at this
suite of characteristics as the property of a syncret-
ic ritual – an overlay on a previous system of ritual:
the formal properties of lunar-solar alignments, the
prehistoric archaeology and the anthropology of the
prehistoric socio-political systems that built the mo-
numents. Since reversal is built into the properties
of a solarised lunar alignment, this is formally con-
sistent with the way in which syncretic systems build
reversed structures on those which preceded them,
and which they are designed to displace, confiscate
and annul. It is also consistent with the continuity
paradigm that Neolithic cosmologies derived from
their antecedent Mesolithic and Palaeolithic forager
ancestors (Silva, Franks 2013). And it is anthropolo-
gically sensitive to the monument builder’s culture.
They were Neolithic cattle herders, who, while conti-
nuing to hunt and forage, had split with their Meso-
lithic/Palaeolithic solely hunter-gatherer origins (Ste-
vens, Fuller 2012). Since hunters earn a wife through
a life-time’s bride-service, and cattle owners purchase
a wife through bride-price payments, this transition
is fraught with dislodging matrilineal rights in the
interests of an emerging patrilineal culture which
prioritises the accumulation of cattle wealth (Sims
2013b). If we pare away the surface overlay that con-
stitutes this solarised lunar syncretism, this formal
exercise requires us to reverse the reversed lunar
phases of lunar standstills while preserving the the-
mes of alternating dark Moons. The result is to rad-
ically simplify the root template out of which lunar-
solar conflation can emerge. Instead of reversed-
phased Moons, we would then have synodic lunar
phases prioritising dark Moon rituals, and instead of
lunar-solar attenuated timescales, we have monthly
alternation every 29/30 days – in short the transit-
ing Moon as seen in the sky by any observer with-
out the intermediary of a monument alignment. We
are left with a ritual system synchronised with lunar
phases which culminate with dark Moon. As a formal
operation, as Palaeolithic continuity and as gender re-
versal, we arrive at a lunar transformational template.
Lunar-phased ritual system considered as a
transformational template
Our interpretation of solarised lunar standstills as
syncretic cosmological systems derives from a prior
condition of Palaeolithic gender equality with (syn-
odic) lunar scheduled rituals. This predicts an initial
situation, cultural origins model of lunar- scheduled
ritual syntax, which entails, consequent upon later
socio-economic changes, later Neolithic lunar-solar
monument alignments. This lunar template suggests
What is a lunar standstill III|
477
that those who monopolise ritual power do so
through a dark Moon performative achievement.
While we have derived this model through the archa-
eoastronomy of solarised lunar monumental align-
ments, it converges with the predictions of sex strike
theory (Knight 1991) that, in conditions of abundant
mega-fauna, monthly dark Moon menstrual seclusion
rituals of matrilineal siblings were used by Palaeoli-
thic women to motivate prospective ‘husbands’ from
other clans for hunting services.
Limiting ourselves in this paper to the archaeoas-
tronomy implications alone, while the model pre-
dicts that the ritual syntax of dark Moon is invariant,
it also predicts that changing relations of power and
economic shifts provide the motivation to transform
this lunar template. For example, matrilineal sibling
coalitions relying on monthly big game hunt provi-
sion will become vulnerable during the mega-fauna
extinctions that took place in Palaeolithic Europe be-
tween the Aurignacian and the Magdalenian (Suro-
vell 2008). Once such resource stress sets in, then
according to regional circumstance, the model pre-
dicts a number of alternative routes for the ritual ap-
propriation of ‘astronomy’. We have found monu-
mental solarised sidereal lunar alignments at Stone-
henge which reverse the synodic Moon while pre-
serving a dark Moon culmination. Ruggles provides
the same data for five other regional groups of mo-
numents in the prehistoric British Isles, and North
showed it for the Avebury monument complex 20
miles north of Stonehenge. Exactly the same ‘astro-
nomy’ is found in the ancestral Hopi ‘Sun Dagger’
site at Fajuda Butte (Sofaer 2012; Sims 2016). It
also allows a solar or stellar cosmology to overlay,
or replace and annul an earlier lunar cosmology.
Evidence for an earlier version of this transforma-
tion can be seen amongst one group of Magdalenien
hunter-gatherers in the Praileaitz I cave in the Basque
country with a solarised and modularised synodic
lunar transformation (Sims, Otero in press). For any
solar cosmology, we would expect it to carry con-
trary evidence of an earlier lunar origin exhibiting
formal attributes according to lunar laws of motion.
For example, it suggests that later versions might
include cosmologies appropriating heliacal risings
and settings as lunar equivalents of waning and
waxing crescent Moons spanning the period of dark
Moon, or using the phases of Venus as lunar equiv-
alents as do the Ona hunter gatherers of Tierra del
Fuego (Bridges 1988). This critique of lunar stand-
stills reveals the prospect of a lunar transformation-
al template being the source of a ‘periodic table’ of
‘astronomies’ used by the world’s cultures through-
out history to buttress ritual systems experiencing
varying degrees of stress and emerging social com-
plexity. Rather than archaeoastronomy becoming a
field discipline for ‘alignment studies’ eschewing in-
terpretation (Ruggles 2011) or as a fringe discipline
(Hutton 2013), it opens up an alternative robust and
scholarly future as cultural astronomy allied to and
embedded within archaeology and anthropology.
Conclusion
The ‘Thom paradigm’ (Aveni 1988) found solstice
and standstill alignments in many prehistoric mon-
uments of NW Europe, and claimed that this was in
the service of eclipse prediction. After fifty years of
research, eclipse prediction has not been accepted
within archaeoastronomy, but the remaining evi-
dence of solar and lunar alignments remain and still
require interpretation. In this paper, we have re-
jected all except one model remaining within archa-
eoastronomy for this evidence. Horizon range, cros-
sover, perturbation and full Moon fail the tests in-
ternal to the discipline to explain the properties of
prehistoric monument alignments on the Moon. Our
critique instead suggests that a model of lunar-solar
conflation points to a ritual syntax centred on dark
Moon rituals is amenable to multiple transformations
consequent upon shifts in socio-political power. If
this model of a lunar transformational template with-
stands refutation, then these examples open the pro-
spect for archaeoastronomy to find a ‘periodic table’
of alignments that will map onto the world’s cosmo-
logies and religions. It predicts that archaeoastron-
omy has a central and essential role to play in future
research into the cosmologies of the world’s cultures.
I would like to thank Fabio Silva, Thomas Gough,
Kim Malville, César González-García and the anony-
mous peer reviewers for their comments on earlier
versions of this paper.
ACKNOWLEDGEMENTS
Lionel Sims
478
story. In I. πprajc, P. Pehani (eds.), Ancient Cosmologies
and Modern Prophets. Slovene Anthropological Society.
Ljubljana: 223–236.
Sims L. 2006. The Solarization of the Moon: Manipulated
Knowledge at Stonehenge. Cambridge Archaeology Jour-
nal 16(2): 191–207.
2007. What is a lunar standstill? Problems of accuracy
and validity in the Thom paradigm. Mediterranean Ar-
chaeology and Archaeometry 6(3): 157–163.
2013a. Where is cultural astronomy going? In F. Pimen-
ta (ed.), Stars and Stones: voyages in archaeoastro-
nomy and cultural astronomy – a meeting of different
worlds. British Archaeological Reports IS 2728. Archaeo-
press. Oxford: 306–310.
2013b. Out of Africa: the solarisation of the Moon. In F.
Pimenta (ed.), Stars and Stones: voyages in archaeo-
astronomy and cultural astronomy – a meeting of dif-
ferent worlds. British Archaeological Reports IS 2728.
Archaeopress. Oxford: 184–190.
2015. 30b – the West Kennet Avenue Stone that Never
Was: Interpretation by Multidisciplinary Triangulation
and Emergence through Four Field Anthropology. In F.
Silva, N. Campion (eds.), Skyscapes: The Role and Im-
portance of the Sky in Archaeology. Oxbow. Oxford:
42–58. 
2016. Forum: What is the Minor Standstill of the Moon?
Journal of Skyscape Archaeology 2(1): 67–76, 95–102.
Sims L., Otero X. in press. Praileaitz I: A Magdalenian lunar-
solar cave at 15,500 BP in the Basque country. Mediter-
ranean Archaeology and Archeometry.
Sofaer A. 2014. Chaco Astronomy. Ocean Tree Books.
Sante Fe.
Stevens C. J., Fuller D. Q. 2012. Did Neolithic farming fail?
The case for a Bronze Age agricultural revolution in the
British Isles. Antiquity 86: 707–722. 
Surovell T. A. 2008. Extinctions of Big Game. In D. Pear-
sall (ed.), Encyclopedia of Archaeology, vol. 2. Academic
Press. Elsevier: 1365–1374.
Thom A. 1971. Megalithic Lunar Observatories. Claren-
don Press. Oxford.
Thomas J. 1999. Understanding the Neolithic. Routledge.
London.
Aveni A. F. 1988. The Thom paradigm in the Americas. In
C. L. N. Ruggles, Records in Stone. Cambridge University
Press. Cambridge: 442–472.
Bridges E. L. 1988. Uttermost Part of the Earth: Indians
of Tierra del Fuego. Dover. New York.
Burl A. 1981. By the light of the cinerary moon: cham-
bered tombs and the astronomy of death. In C. L. N. Rug-
gles, A. W. Whittle (eds.), Astronomy and Society in Bri-
tain During the Period 4000–1500 BC. British Archaeo-
logical Reports BS 88. Archaeopress. Oxford: 243–274.
González-García C. 2016. Reply: Lunar extremes, lunar
cycles and the minor standstill. In L. Sims, C. González-
García and J. McKim Malville, Forum: What is the minor
standstill of the Moon? Journal of Skyscape Archaeology
1(3): 67–102.
Heggie D. 1981. Megalithic Science: Ancient Mathema-
tics and Astronomy in North-west Europe. Thames and
Hudson. London.
Hutton R. 2013. The Strange History of British Archaeoas-
tronomy. Journal for the Study of Religion, Nature and
Culture 7(4): 376–396.
Knight C. 1991. Blood Relations: Menstruation and the
Origins of Culture. Yale. London.
Morrison L. V. 1980. On the analysis of megalithic lunar
sightlines in Scotland. Archaeoastronomy 2: 65–77.
North J. 1996. Stonehenge: Neolithic Man and the Cos-
mos. Harper Collins. London.
Parker Pearson M. 2012. Stonehenge. Simon and Schus-
ter. London.
Ruggles C. L. N. 1999. Astronomy in Prehistoric Britain
and Ireland. Yale. London.
2011. Pushing back the frontiers or still running around
the same circles? ‘Interpretive archaeoastronomy’ thirty
years on. In C. Ruggles (ed.), Archaeoastronomy and
Ethnoastonomy: Building Bridges Between Cultures.
University Press. Cambridge: 1–18.
Silva F., Pimenta F. 2012. The Crossover of the Sun and
the Moon. Journal of the History of Astronomy 43(2):
191–208.
Silva F., Franks R. 2013. Deconstructing the Neolithic Myth:
The Implications of Continuity for European Late Prehi-
References