Ancient Zodiacs, Star Names, and Constellations: Essays and Critiques
Some Critical Comments on "Origins of the Ancient Constellations" by John Rogers by Gary D. Thompson
Copyright © 2010-2018 by Gary D. Thompson
Return To Site Contents Page
Some Critical Comments on "Origins of the Ancient Constellations" by John Rogers
The two-part article by Dr John Rogers, Jupiter Section Director of the British Astronomical Association, "Origins of the ancient constellations: I. The Mesopotamian traditions." (Journal of the British Astronomical Association, Volume 108, 1998, Number 1, February, Pages 9-28); and "Origins of the ancient constellations: II. The Mediterranean traditions." (Journal of the British Astronomical Association, Volume 108, 1998, Number 2, April, Pages 79-89) tends to be highly regarded. However, both parts of this article (comprising a total of 31 pages) needs to be used with caution. The article, the result of prolonged research, ultimately comprises a speculative and misleading synthesis compiled in part from dated and/or unreliable sources. These sources include: Richard Allen, Robert Brown, Andrew Crommelin, Alex Gurshtein, Willy Hartner, Edward Maunder, Michael Ovenden, Werner Papke, Archibald Roy, Richard Proctor, Giuseppe Sesti, and David Ulansey. These authors have had a major influence on the ideas expressed in the article.
Origins of the ancient constellations: I. The Mesopotamian traditions.
The standard source on Babylonian star-names and constellations (and associations) still remains Planetarium Babylonicum by Felix Gössmann (1950). Unfortunately there is no English-language edition. Gössmann cites some 124 sources for his compilation. Whilst the book is now 60 years old forty percent of its sources are earlier than 1920 (i.e., are now over 90 years old). There has been an enormous increase in our knowledge of cuneiform philology in the past 60 years and Gössmann's compilation has some problems regarding the correct rendering/identification of constellation/star names.
The author does not provide any directly supporting statements contained in either mythological or star-list texts for his intriguing theory of two overlapping traditions of constellations - a divine/heraldic tradition and a rustic/farming tradition. He can only argue the concept is implied in the star-list material. In the end the concept is a subjective judgment. (Most of the Mul.Apin stars/constellations are associated with gods/goddesses.) The fact the author allows that many constellations belong to both traditions weakens this theory. What also weakens this theory is the overall scheme for the placement of the constellations in the paths of 3 great gods, Ea, Anu, and Enlil. This scheme of assigning the constellations is explicit within multiple texts from the Boghazköy star-list, the Astrolabe tradition, and the Stars of Elam, Akkad, and Amurru list onwards. Whether there was more than one (type of) tradition of constellation/star names in Mesopotamia see: "Mesopotamian Star Lists." by Wayne Horowitz. In: Handbook of Archaeoastronomy and Ethnoastronomy edited by Clive Ruggles (2015; Chapter 168 (but also given as 187), Pages 1829-1833). "Sumerian and Akkadian names of stars and constellations occur in cuneiform texts for over 2,000 years, from the third millennium BC down to the death of cuneiform in the early first millennium AD, but no fully comprehensive list was ever compiled in antiquity. Lists of stars and constellations are available in both the lexical tradition [dictionary lists] and astronomical-astrological tradition of the cuneiform scribes. The longest list in the former is that in the series Urra = hubullu, in the latter, those in Mul-Apin."
Pages 9 & 10:
The so-called (1) Early pictographic phase, circa 3200-2100 BCE; and (2) Boundary-stone pictographic phase, circa 1350-1000 BCE, are essentially baseless speculations arising from the unfounded viewpoints of Willy Hartner and Alex Gurshtein. The theme being offered is at least some of the zodiacal constellations came into existence at a quite early date and had a functional significance. From the ideas of Willy Hartner a supposed starting point for the constellations/zodiac is obtained and from the ideas of Alex Gurshtein developmental steps for the constellations/zodiac are obtained. Without their ideas to focus on there is no concept of "the earliest zodiacal constellations" - there are simply constellations until circa 700 BCE when the Babylonians moved to created a 12-constellation zodiac.
It not established that constellations or constellation symbols are being depicted on cylinder seal iconography and similar from the Sumerian and Akkadian Period circa 3200-2000 BCE. During the pioneering period of assyriology it was believed that Bull, Lion, Scorpion, Water-Carrier, Swallow/Field, Hired Man, and Goat-Fish iconography from this period were representations of the zodiacal constellations.
It not established that constellations or constellation symbols are being depicted on Babylonian boundary-stone (kudurru) iconography from the Kassite Period circa 1530-1160 BCE. It was the pioneering British assyriologist Stephen Langdon who promulgated this idea. He incorrectly believed that Bull, Lion, Furrow, Scorpion-Archer, Hired Man, and Goat-Fish iconography on boundary-stones from this period were representations of the zodiacal constellations. It is established that god/goddess symbols are depicted. In the early period of Assyriology it was common to identify these symbols as depictions of the zodiacal constellations. Further work in Assyriology has changed this assumption. It not established that constellations or constellation symbols are being depicted. It is established, however, that god/goddess symbols are depicted. For a recent attempt to establish the astral nature of kudurru symbols (from the Cassite Period, circa 1530-1160 BCE) see: "Eine neue Interpretation der Kudurru-Symbole," by Ulla Koch, Joachim Schaper, Susanne Fischer, and Michael Wegelin (Archive for History of Exact Sciences, Volume 41, 1990/1991, Pages 93-114). However, the attempts to date kudurru by assuming their iconography has astral significance and then using the arrangement of their iconography to establish astronomical dates is both speculative and unproven. Ursula Seidl, a present-day kudurru expert, maintains in her article "Göttersymbole und -attribute." (Reallexikon der Assyriologie (Dritter Band 3, 1957-1971, Pages 483-490)) that kudurru iconography has no astral significance. (See also her book: Die Babylonischen Kudurru-Reliefs Symbole Mesopotamischer Gottheiten (1989). In this book, regarded as the standard study of kudurru iconography, she maintains her scepticism that kudurru symbols have an astral significance.)
These statements reflect the ideas of Willy Hartner (an early set of 4 constellations on the ecliptic marking the tropical points) and Alexander Gurshtein (a gradualist concept of the development of the constellations and the zodiac). Hartner's conjecture of an early constellation set marking the 4 tropical points and Gurshtein's ideas of a gradualist development of the zodiacal constellations both conflict with the cuneiform evidence from Mesopotamia. In her 2005 thesis The Lion-Bull Combat as an Astronomical Symbol in the Context of the Origin of the Constellations Kate Anscombe supports the views of Willy Hartner on the earliest Mesopotamian constellations and claims to identify slight time-related changes (precessional changes) in lion-bull iconography. Unfortunately she relies exclusively on secondary sources and surprisingly no recalculated "accuracy test" of Hartner's calculations is made. Assyriologists have not accepted Hartner's conjectures on early iconography as representations of constellations.
The zodiac was a development from the Babylonian scheme of 17/18 constellations/stars marking the path of the moon. The Babylonian system of 17/18 constellations/stars zodiac marking the path of the Moon belongs to the Assyrian Period (and perhaps originated circa 1000 BCE) and was still in use in the 7th-century BCE and contained the constellations that were to form the 12-constellation solar zodiac. At least 5 of these 17/18 constellations/stars are not previously listed but are additional constellations/named stars in the Mul.Apin series. Prior to the Mul.Apin system of 17/18 constellations/stars (circa 1000 BCE earliest) the ecliptic was not specifically marked (or identified) in Babylonian astronomy. (Not all the 17/18 constellations/stars that were used to mark the path of the Moon were within the ecliptic.) Earlier than circa 1000 BCE there was, in Babylonian astronomy, no intentional system of marking the path of the ecliptic.
The term "proto-zodiacal" is somewhat misleading. Some constellations that later formed part of the zodiac were established in Mesopotamia circa 2000 BCE or perhaps earlier, and some were perhaps originally used as seasonal markers. However, in the early 2nd millennium BCE these constellations formed part of the Babylonian system of "three stars each" i.e., the Paths of Anu, Enlil, and Ea. There is no mention of the zodiacal scheme in Babylonia, or elsewhere in the Occident, prior to the 1st millennium BCE. The evidence clearly shows the zodiacal scheme originated in Babylonia. The astronomy of the Mul.Apin series established the preconditions for the establishment of the zodiac. The zodiac did not originate in either Elam or Sumer. We know nothing about Elamite or Sumerian astronomy - if either indeed existed. There is actually greater likelihood that the earliest constellations originated in the southern Levant (basically encompassing the territory of modern-day Israel and Jordan) during the Chalcolithic Period/early Bronze Age circa 4500-2000 BCE.
The expression "Stars of Elam, Akkad, and Amurru" perhaps suggests that, even very early, at least in Mesopotamia, the constellations originated as independent formal schemes having a calendrical purpose. The names Elam, Akkad, and Amurru reflect the political situation in Old Babylonian times. This enables the assumption that the lists are old and come from the Old Babylonian Period. However, the assyriologist Wayne Horowitz cautions the lists of Elam-, Akkad-, and Amurru- stars may not be as old as the Old Babylonian period as suggested by Bartel van der Waerden.
It certainly has not been demonstrated that the earliest astronomical iconography originated in Elam. Hartner did not track the origin of the lion-bull symbol to Elam circa 4000 BCE. Hartner's earliest convincing lion-bull iconographic evidence (on a pitcher from Uruk) originated some 700 years later than the 4000 BCE seasonal marker date he identified. We do not know the culture responsible for originating the lion-bull combat/contest symbolism. (Interestingly, the Ibex (postulated by Hartner as also depicting an early cardinal constellation) appears on prehistoric pottery of the Near East dating back to to circa 5500 BCE.) If a precise celestial event is being represented then more attention surely should be given to individual stars rather than the focus on large constellations with undefined boundaries (and separated by one-quarter of the sky). It has been pointed out that with Hartner's method there is a large degree of imprecision with geographical latitude and epoch regarding the concept of early large constellations. This looseness is heightened when the certainty of which culture originated the constellations hypothesized is not satisfactorily established. The criticism made is that substantial changes to latitude appreciably change the hours between settings (i.e., interval relationships) of the constellations.
Origins of the ancient constellations: II. The Mediterranean traditions.
There was no "zodiac tradition" that developed from 3200-500 BCE. The constellations and their stars - and their uses - developed from circa the Old Babylonian period to the Seleucid period. The Mesopotamian constellations were largely developed and consolidated during the latter half of the 2nd-millennium BCE.
It is clear that the Babylonians had 2 serpent constellations - both at least by the Old Babylonian period. In the 10-star version of The Prayer to the Gods of the Night (AO 6769, dated to the Old Babylonian period) the Dragon and also the Serpent are 2 of the constellations listed. The term mušhuššum (Akkadian loan-word of Sumerian MUŠ.HUŠ) (mušhuššum-dragon) was a "furious snake (serpent)"/"aweful snake (serpent)." The god Marduk was associated with snakes and the mythical Dragon (snake-dragon/serpent-dragon (horned serpent) was the sacred animal (symbol) of the god Marduk. It is presently an unidentified Mesopotamian constellation (= the large circumpolar constellation Draco?). The classicist David Kidd states (Aratus Phaenomena (1997, Page 192): "It [the Dragon] probably came to the Greeks from the Babylonians." The symbolism of Babylonian dragons is different to that of snakes. The term bašmum was a mythical poisonous snake, a horned viper (a kind of serpent-dragon or snake-dragon) possessing several tongues and jaws. The Horned-Serpent was the sacred animal of Marduk. It is provisionally identified as a constellation north of Libra but also provisionally identified with Hydra (a very long constellation - also the largest in the sky - located in the region of the celestial equator). Regarding the Greek Serpent constellation the classicist David Kidd states (Aratus Phaenomena (1997, Page 206): "The origin is unknown: perhaps the basic group was the snake, and the holder added to it by the Greeks."
The idea of "Mediterranean navigator constellations" as a substantial source for some of the Greek constellations remains an unsupported supposition. Likewise the conjectural date of circa 2800 BCE for the origin these supposed "Mediterranean navigator constellations" remain an unsupported supposition. The more recent studies by the American astronomer Brad Schaefer have undermined the "void space" and "constellation alignment" ideas promulgated by the astronomers Andrew Crommelin (British comet specialist), Edward Maunder (British sun-spot specialist), Michael Ovenden (Canadian astro-physicist), and Archibald Roy (British astro-physicist). Schaefer's recent constellation studies have also undermined the Minoan theory of the source of Greek constellations. Additionally, we have no actual knowledge (evidence) of any Minoan constellations.
Evidence for direct constellation borrowing by the Greeks from Mesopotamia is almost non-existent. During the Hellenistic period it is possible that Berossus and some Chaldaean contemporaries made the Babylonian sphaera familiar to the Greeks. It is possible that Babylonian uranography was passed to the Greeks through particular intermediaries such as the Phoenicians and Egyptians. There were likely "competitive" schemes of Greek sphaera until the wide adoption of the sphaera developed by the Greek astronomer Eudoxus of Cnido and diffused through his written works on the constellations (4th-century BCE). The ultimate success of the Sphaera Graecanica as we have it today (i.e., its complete acceptance by the Greek world and later the Roman world) was largely due to the work of the Greek astronomer Eudoxus of Cnidus (4th-century BCE) and the Greek poet Aratus of Soli (3rd-century BCE). Eudoxus constellated and catalogued the entire Greek sky in his works Enoptron and Phaenomena. Aratus later turned these works into an astronomical poem concerning the constellations. The Phaenomena became hugely popular in the Graeco-Roman world. Without this popularisation by Aratus the works of Eudoxus may never have exerted the lasting influence they achieved.
Rogers' position is only the zodiacal and a few related constellations were inherited by the Greeks from the Babylonians. He believes the remaining Mesopotamian constellations are different from the Greek ones: different groups of stars, unrelated names, different myths. To a degree the origin of the non-zodiacal Greek constellations remains uncertain. Books/articles written in the late 19th-century or early 20th-century – by particular persons (= particular amateurs) - have perhaps tended to overestimate the number of non-zodiacal Greek constellations identified in the Mesopotamian records. It can be easily argued - and has been - that considerably more Babylonian constellations were adopted by the ancient Greeks. Strangely, the Serpent is included even though the author has only a "few" paragraphs earlier denied it was a Babylonian constellation and declares it to likely to have been adopted from its origin as a "Mediterranean navigator constellations." Congruence between Babylonian and Greek constellations/conceptions of constellations and between constellation names/types include (zodiacal constellations excluded): the Twins Babylonia mas-tab-ba-gal-gal-la = the Great Twins, α and β Geminorum, the stars Castor and Pollux), a Dragon, a Snake, a Raven, an Eagle, a Fish, Orion the hunter (Babylonia sipa-zi-an-na (siba-zi-an-na) = True Shepherd of Anu/Heaven), and the Greek constellation Deltotan (= the Babylonian triangle of stars gis apin (the Plough)). Triangulum was one of the 48 constellations listed by the 2nd-century astronomer Claudius Ptolemy. This is a total of 8 constellations (zodiacal constellations excluded). Due to the limited nature of the evidence it has not been resolved whether these are borrowings or coincidences, or a mixture of both (with adaptations in some cases).
I would offer that most of the views of Richard Allen and Robert Brown Junior on the earliest origins of the Western constellations are inaccurate. Neither person was professionally qualified to undertake the study of the subject matter (Allen was a businessman, and Brown was a solicitor). Giuseppe Sesti's book The Glorious Constellations is an insubstantial and unreliable "coffee table" book based on unreliable and outdated sources. Sesti's book should not be used. The viewpoints of Richard Allen in Star-Names and Their Meanings, and Robert Brown Junior in Researches Into the Origin of the Primitive Constellations, are thoroughly outdated and misleading, and best ignored.
There are 2 flawed concepts here: (1) the constellation alignment argument, and (2) the "void zone" argument. Regarding the constellation alignment concept. It was argued by Richard Proctor, Edward Maunder, and Andrew Crommelin that the original constellations were aligned to the celestial equator and the colures. According to Michael Ovenden the ancient constellations show a rough symmetry around the star alpha Draconis (now about 25 degrees from Polaris). This is a version of the constellation alignment argument known as the "polar alignment" argument. Since alpha Draconis was the star of the north pole circa 2600 BCE, Ovenden concluded that most constellations were designed circa 2600 BCE (plus/minus 800 years). There is, however, no evidence that people in the 3rd-millennium BCE had, or used, this type of astronomical positional system. There is no compelling reason to believe that the constellations were specifically established, at a particular time and place, as a system of coordinates. The scheme proposed by Ovenden (a version of the Minoan origin argument) does not apparently try to separate the Greek constellations known prior to Eudoxus (i.e., the earlier Greek constellations of Homer and Hesiod) or the later Greek constellations that are distinctly Babylonian in origin (i.e., the zodiacal constellations). These types of arguments are highly speculative and entirely deductive and fixed on a priori assumptions.
The "void zone" method, though popular since its popularisation by Edward Maunder, has multiple problems. The chief premise of the "void zone" is that the classical Greek constellations (i.e., the Aratean constellations) were designed at one definite time and in one place, according to a preconceived plan. The argument for establishing the time and place of the Aratean constellations is based on the extent of the vacant space left around the south pole of the celestial sphere when all but the Aratean constellations are removed; and the apparent movement of the stars due to precession. The further assumption made is that the area of the globe that was not constellated in the description of Aratus was centred on the south celestial pole at the date when the constellations were fixed. The size of the "void zone" is taken as a clue to the latitude at which the constellation inventors lived. A date is found when, by allowing for precession, the centre of the "void zone" on the globe is in the position of the south celestial pole. The "void zone" method has too many significant flaws. The subjectivity of the method is demonstrated by the varying estimates of the radius of the "void zone" (30 degrees to 40 degrees) and the varying estimates of the date of origin given by precession (1400-2800 BCE). Anyway the boundaries of the "void zone" cannot be accurately defined as we lack the understanding of the original boundaries of the classical Greek constellation figures. Due to our lack of knowledge of the boundaries of the Aratean constellations the "void zone" method is inherently subjective and its use can lead to no real agreement (as it has failed to do) regarding the latitude and date for the constellations being designed at one definite time and place.
Many of the Aratean constellations show a similarity with Babylonian constellations. The Greek constellation scheme of Aratus of Soli (3rd-century BCE) contains a mix of both Babylonian constellations and non-Babylonian constellations. The Babylonian component of the Aratean constellations is traceable to both Babylonian "star calendar" constellations of the 2nd millennium BCE and also to Babylonian constellations listed in the later Mul.Apin series (circa 1000 BCE). (The few known 8th-century BCE constellations of Homer mirror the constellations already existing in the Babylonian scheme.) The Babylonian scheme of constellations has always been a mix of constellations mentioned by Aratus and other constellations outside the Aratean scheme. A definite Babylonian influence on the later Greek scheme of constellations is reasonably indicated. It is obvious that the Greeks borrowed certain constellations from the Babylonians and it is obvious that the constellations could not have originated, or been adopted, as a single devised scheme by either the Babylonians or the Greeks. If the constellations originated as a set circa 2000-2800, as commonly claimed by the proponents of the "void zone" method, then they cannot have originated with the Greeks. However, the latitude at which the constellations were believed to have originated as a single scheme cannot refer to Mesopotamia because their earliest scheme of constellations, though dating to the 2nd millennium BCE, was a mix of constellations mentioned by Aratus and other constellations outside his scheme.
Crediting the Minoans, as some like to do, as the makers of the classical constellations and offering explanations based on the destruction of Minoan civilization and the later ineptitude of the Greeks as observers are also not convincing. There is no evidence that the classical Greek scheme of constellations existed anywhere prior to its evolvement in Greece circa 500 BCE. This includes the fact that there is no evidence that the particular Greek scheme of 12 zodiacal constellations existed anywhere prior to its evolvement in Greece circa 500 BCE. The difficulty with maintaining an ancient zodiac is how can a late Mesopotamian zodiac (developed circa 500 BCE) and comprised of 12 constellations (and 12 equal divisions), and substantially borrowed by the Greeks, have been in use by anybody hundreds of years earlier. (Or even thousands of years earlier, prior to the existence of the Babylonian civilization which demonstrably created it.) The flawed "void zone" argument has become a common tool for maintaining that a Neolithic zodiac (and fully constellated sky) can be reasonably be proposed. The "void zone" argument can hardly substitute for the lack of clear evidence (which tends to fall under the murky heading of "tradition"). Even if the "void zone" argument were correct it has never offered support for the idea that the constellations could have existed as a deliberately planned set extending back some 6000-8000 years BCE (or further). The use of the "void zone" argument controls the feasible range for the dating of the constellations if they are considered to have originated as a deliberately planned scheme. Interestingly, Edward Maunder, a committed proponent of the "void zone" argument, in his later articles on the topic attempted to overcome this limitation by implying a very slow developmental period for the final scheme of constellation design (see: "Origin of the Constellations", The Observatory, Volume 36, 1913, Page 330).
The version of the arguments that were promulgated by Michael Ovenden and Archibald Roy remain highly speculative and highly unlikely. They have been effectively demolished by the more recent and more stringent investigations of the American astronomer Brad Schaefer who has convincingly traced many Greek constellations and Greek sky-lore back to Mesopotamia and the Mul.Apin star catalogue circa 1300 BCE.
Also, see the important critical papers: "Statistical Dating of the Phenomena of Eudoxus." by Dennis Duke (DIO-The International Journal of Scientific History, Volume 15, December 2008, Pages 7-23). (URL: http://people.sc.fsu.edu/~dduke/eudoxus.pdf). And, "A 'Watermark' of Eudoxan Astronomy." by Elly Dekker (Journal for the History of Astronomy, Volume XXXXIX, Part 2, Number 135, 2008, Pages 213-238). (URL: http://articles.adsabs.harvard.edu/full/2008JHA....39..213D) (Note: Both are important articles for correcting the mistaken notion that Eudoxus was referring to constellations established circa 1000 BCE, and not in his own lifetime circa 370 BCE.)
Nothing the author has proposed as evidence makes a convincing case that the classical constellations of the zodiac and some others developed progressively from the 4th- to the 1st-millennium BCE. A developed Mesopotamian constellation set is a 'rapid' development of the late 2nd-millennium BCE and is closely associated with the needs for observing and locating celestial omens. The concept of a zodiac - and the origin of constellations that would latter form the constellations of the zodiac - was very much tied to the 1st-millennium BCE. At least 20 of the classical constellations came from Mesopotamia. Other classical constellations reflect the influence of Greek mythological traditions - resulting in multiple constellations groups connected by storyline. Two constellation groups comprising a common mythological theme are: (1) the 5 constellations of the Peseus-Andromeda group: Perseus, Andromeda, Cassiopeia, Cepheus, and Cetus; and (2) the 5 constellations mythologically connected with Orion: Scorpio, Canis Major, Canis Minor, Lepus, and the Pleiades. Also, a 3rd group of constellations that are mythologically associated are the 4 constellations: Argo, Centaurus, Ara, and Corvus. The Greeks essentially mapped the whole sky in mythological terms. (See: Star Myths of the Greeks and Romans (1997) by Theony Condos.)
Appendix 1: Precessional Argument
There are 4 basic methods involved with the precessional argument for dating the Western constellations: (1) The constellations marking the ecliptic of in Aratean lore. (2) The constellations marking the 11 "heavenly circles" in Aratean lore. (3) The unconstellated space around the south celestial pole (void zone) based on Aratean lore. (4) The orientation of the constellations to the north celestial pole based on Aratean lore.
Appendix 2: Summary of Arguments Against the Void Zone Hypothesis
The proponents of the void zone argument generally take their own work as being the only method to obtain a dating solution and ignore all the other results from philology, archaeology, history, and mythology. These other sources of information show the first surviving evidences for constellations to be substantially later than the void zone proponents would allow. There are a number of substantial arguments against the ‘void zone’ hypothesis.
(1) It is not established that we know all the ancient constellations known to the Greeks. We do not know whether the Aratean list of constellations is complete; especially regarding southern constellations. The existence of other southern constellations (further south) omitted by Aratus could affect the determination of the centre of the ‘void zone’ and estimated epoch, and observer latitude, which could be more southernly.
(2) It is not known what stars comprised the boundaries of the ancient constellations. If assumptions of the southern edge of the southernly constellations are incorrect then this will lead to errors with date and latitude. It is not known whether the constellation boundaries were inconsistent (liable to change). The earliest knowledge of specific stars comprising most of the constellation figures only goes back to Hipparchus' Commentary on Aratus. We do know that the Greeks did change the boundaries of the Aratean constellations. Aratus also changed the boundaries of several of the Eudoxan constellations.
(3) Proponents of the ‘void zone’ hypothesis make the assumption that the constellations (or at least the most southern constellations were all developed at approximately the same time and place. However, there is no evidence or particular reasons to believe this assumption. The incorrectness of this 'void zone' argument will lead to errors in the determination of the centre of the 'void zone' (= errors in date and latitude). Any argument that the 'void zone' itself provides evidence that the southern constellations were very old in Aratus' time is simply a deduction that is lacking supportive evidence. (Even Maunder realised the inherent problems of arguing for an early static constellation set.)
(4) Proponents of the use of the 'void zone' hypothesis generally overemphasize this approach as being a simple, direct, and reliable methods to achieve a solution for the date of the constellations. Ignored are the results of other methods such as history, mythology, archaeology, and philology. These other methods combine to indicate that the surviving evidence for the Western constellations show them to be established together in Greek the Greek sky much later than the use of the 'void zone' method indicates.
(5) A now standard question by ‘void zone’ proponents is: Why wasn't the constellation lore inherited in Aratus' Phainomena updated - especially by navigators - as it became obvious that it was in error (i.e., mismatched the visible sky)? This question is overlooked as an argument against the validity of the 'void zone' argument. Commonly invoked is the supposed existence of an inherited archaic constellation globe and a separated nautical tradition. Both without evidence. However, even 'void zone' proponents have to acknowledge that Hipparchus did identify problems with Aratus' Phainomena and update astronomical statements in the poem. In his Commentary on Aratus, Hipparchus also acknowledged the substantial descriptive correctness of Aratus' Phainomena. Many of the criticisms of Hipparchus are unrelated to the positions of the constellations with regard to the celestial equator. Because of the limited amount of information that has come down to the present-day it is not possible to reliably know whether problems went unnoticed earlier or if there were attempts to update the constellation lore that are now lost because they were not successfully adopted into the main body of constellation lore.
(6) There is no realistic evidence that an Aratean scheme of constellations existed earlier than the time of Eudoxus of Cnidus (4th-century BCE). To argue that the evidence is now lost or not yet recovered is to sidestep the particular difficulty with speculation. This does not automatically mean that the first surviving mention of a fully constellated Greek sky is also an argument for the approximate date of this particular development. Some constellation names and figure are met with earlier in Greek sources but these named constellations are nowhere near the detailed list given by Aratus.
(7) We do not know whether the person(s) engaged in the development of the inherited Western constellations constructed the constellation scheme all the way to the southern horizon. If this was not done then this places the deduced latitude farther south than the estimates which have been given. We know the Greeks didn't have constellations all the way to the southern horizon.
(8) I cannot recall any 'void zone' proponent who has included discussion of the very real issues of atmospheric extinction and refraction in their method for calculating a date. Even 1st magnitude stars cannot be seen to the horizon. The effects of atmospheric extinction and refraction would place the deduced latitude of the constellation developer(s) farther south than the estimates which have been given.
The void zone proponents assume that their hypothetical early constellation set, once originated, was not updated. Indeed it was transmitted unchanged - with the aid of a hypothetical star globe - to Eudoxus. However, we have the Greek example of Hipparchus - circa 100 years after Aratus - identifying problems with the celestial sphere described by Aratus in Phainomena. For the most part, Hipparchus was in agreement with the celestial sphere described by Aratus. This point is generally overlooked by void zone proponents.
It is indicated that the most reasonable hypothesis for the origin of the main group of the Greek constellations is they had multiple sources and were made over a wide range in time (stretching mainly from circa 2000 BCE to circa 400 BCE. Some were developed circa the time of Eudoxus (4th-century BCE) and some were borrowed from earlier scheme of Babylonian uranography which was consolidated circa the late 2nd-millennium BCE. The Greek zodiac originated in Babylonia with what was a recently finalised scheme. The Greek consolidation of the 2 streams of constellation development was effectively achieved circa the 4th-century BCE (with the impetus of the introduction into Greece of the Babylonian zodiacal constellations). It is indicated that the earlier scheme of Babylonian uranography remained largely unknown to the Greeks until circa 800 BCE. There is no satisfactory evidence for the long popular hypothesis that many of the western constellations are very ancient, originating long before the 3rd-millennium BCE. Alongside this idea there is no satisfactory evidence for the long popular hypothesis that the main Western constellations were primarily developed in a single epoch/time circa in the mid 3rd-millennium BCE. There is certainly no support for the hypothesis that most, if not all, of the Eudoxan-Aratean constellations had only been developed circa 500 BCE. This idea is convincingly contradicted by the evidence from Babylonian uranography. There is no reason to 'mix-and-match' between these hypotheses - specifically the inclusion that the main Western constellations were primarily developed as part of a constellation set circa 3rd-millennium BCE.
Appendix 3: The Date of Aratus' Equator
Mary Evershed - a critic of the 'void zone' arguments - concluded that the date of Aratus' equator, though it will not fit any date, but then concluded the date circa 800 BCE agrees quite well with Aratus' description. This estimation by Evershed conflicts directly with other claims - for example by Archibald Roy and Goran Henriksson - that the date of Aratus' equator is circa 2500 BCE (Robert Brown Junior asserted 2084 BCE). The wide variation in dates from the same description of constellations along the celestial equator given by Aratus in Phainomena is evidence of a large degree of uncertainty. Evershed also points out that “the two tropical circles; compared with the actual circles of any date ... are in error in one place or another.” The French mathematician Jean-Baptiste Delambre also examined the issue of dating the celestial equator described by Aratus and concluded it was impossible to come to any decision. (See: Histoire de l'astronomie ancienne (1817, 2 Volumes).)
Appendix 4: Two Critical Arguments Against the Void zone Approach and the Constellation Set Approach
The "void space" argument is a simplistic substitution for the more rigorous application of historical evidence (i.e., extent cuneiform and classical texts, philological analysis of constellation names, and constellation iconography and mythology). The 200 year-old "void space" argument originated at a time when philology and archaeology were both under-developed and unable to be applied in any meaningful way. The main premise of the "void zone" argument is that the Classical constellations were designed at one definite time and in one place, according to a preconceived plan. The argument by Michael Ovenden (a modern proponent of the "void zone" argument) for establishing the time and place of the origin of the Aratean constellations is based on the extent of the vacant space left around the south pole of the celestial sphere when all but the Aratean constellations are removed (that is, when the post Aratean constellations are removed); and the apparent movement of the stars due to precession is taken into account. The further assumption made is that the area of the celestial globe that was not constellated in the description of Aratus was centred on the south celestial pole at the date when the constellations were fixed. The size of the "void zone" is taken as a clue to the latitude at which the constellation inventors lived. A date is found when, by allowing for precession, the centre of the "void zone" on the celestial globe in in the position of the south celestial pole. The subjectivity of the method is demonstrated by the varying estimates of the radius of the "void zone" (30 degrees to 40 degrees) and the varying estimates of the date of origin given by precession (1400 BCE to 2800 BCE with ± varying between 200 years and 800 years). Anyway, the boundaries of the "void zone" cannot be accurately defined as we lack the understanding of the original boundaries of the Classical constellation figures.
The importance of the ecliptic and the development of the equally divided 12-constellation zodiac does not appear until after the start of the Persian Period in Mesopotamia (circa 500 BCE). The evidence indicates that it was the astronomy of the Babylonian Mul.Apin scheme (circa 1000 BCE) that established the preconditions for the importance of the ecliptic and the establishment of the Babylonian zodiacal scheme which was later adopted by the Greeks. The Babylonian scheme of 12 zodiacal constellations was derived from a system of 18 constellations (established during the Assyrian Period, starting circa 1100 BCE) along the ecliptic to mark the path of the moon. The question remains how can a late Babylonian zodiac (developed circa 450 BCE) comprised of 12 constellations (and 12 equal divisions) have been in use by Homer some 300 years earlier? (And also have had an even earlier origin circa 8000 BCE - which is well prior to the existence of both the Sumerian and Babylonian civilizations.) This point can be expanded further. Babylonian constellations of the 2nd-millennium BCE appear in the Greek constellation scheme of Aratus. (The few known constellations of Homer mirror the constellations already existing in the Babylonian scheme.) The Aratean constellations also show a similarity with the later Babylonian constellations listed in Mul.Apin. A definite Babylonian influence on the Greek scheme of constellations is reasonably indicated. If the constellations originated as a set circa 2000-2800 BCE then they cannot have originated with the Greeks. However, the latitude at which the constellations were believed to have originated as a single scheme cannot refer to Mesopotamia because at this time their scheme of constellations was a mix of constellations mentioned by Aratus and other constellations outside this scheme. Crediting the Minoans, as Michael Ovenden did, as the makers of the Classical constellations and offering explanations based on the destruction of Minoan civilisation and the later ineptitude of the Greeks as observers are not convincing.
Return to top of page.
This web page was last updated on: Sunday, November 5, 2017, 8.00 am.
This web page was created using Arachnophilia 4.0 and FrontPage 2003.
You can reach me here by email (but first delete the obvious attempt in the email address to foil the spammers):
Return To Site Contents Page