The literature

This page exists so that nobody has to take the article on trust.  Every source it rests on is listed here with what it actually measured, who wrote it, and which way it cuts.

It is deliberately not a reading list of things that agree with this site.  The source that does the most damage to the "fully human population" reading is Zollikofer 2024, an open-access Nature paper by the Dmanisi excavators.  The source that treats the creationist position most honestly is Line 2020, a Journal of Creation paper that concedes pathology cannot explain the small brains and that the cause is unknown.  Both are tagged by what their data show, not by what their authors believe.

How to read an entry

Measurement or model.  A CT volume, a synchrotron count of growth lines, or an argon plateau is an observation.  A stature estimate built on an assumed growth curve is a model.  Both are useful and they are not the same kind of thing.

Direction.  Assigned by what the source shows, not by what its author believes.

Verified.  Whether this project read the text directly (full), only its abstract (abstract), or only through another source that quotes it (via).  Nothing is quoted in the main article unless it is marked full or abstract, and every quotation was read out of downloaded bytes.  Thirteen of the thirty sources are abstract only, and the article attributes nothing to them beyond their published abstracts.  Every abstract was re-fetched as bytes during the verification pass of 6 September 2026, from NCBI E-utilities or the OpenAlex Crossref deposit.

Field definitions

Coverage and its limits, stated plainly.  This catalogue covers the primary descriptions and dates for Dmanisi, Nariokotome, Java, Zhoukoudian and Ngandong, the four behavioural finds most often cited, the three developmental studies, and the creationist treatments from Lubenow 2004 to Line 2020 and Tomkins 2019.  Three items could not be obtained and are listed at the bottom.  The most important missing primary is the Nariokotome monograph.

The specimens and the range

A Complete Skull from Dmanisi, Georgia, and the Evolutionary Biology of Early Homo

Lordkipanidze, David, et al.  Science 342(6156):326–331, 2013.  https://doi.org/10.1126/science.1238484

camp: mainstream topic: specimens method: measurement direction: complicates verified: full

Key figures: Skull 5 endocranial volume 546 ± 5 cm³ by CT;  Dmanisi range 546–730 cm³;  stature 146–166 cm, body mass 47–50 kg;  encephalization quotient about 2.4, inside the Australopithecus range.

The single most useful paper in the file, and it cuts both ways.  It supports the creationist habit of reading morphological spread as within-population variation, in the excavators' own words, and it retracts the species name H. georgicus.  It undercuts the young-earth reading in the same breath, because the population whose variation is defended has adult brains of 546 to 730 cm³ and an EQ in the australopith range, and the comparison drawn is with chimpanzee demes, not human populations.  Read from an open author-quality copy at hesp.irmacs.sfu.ca with the printed folios intact.

Natural History of Homo erectus

Antón, Susan C.  Yearbook of Physical Anthropology 46:126–170, 2003.  https://doi.org/10.1002/ajpa.10399

camp: mainstream topic: specimens method: review direction: complicates verified: full

Key figures: Table 3 cranial capacities from D2700 at ~600 cc to Ngandong 6 at 1,251 cc;  brain-size trend in Asian erectus of about 160 cc per million years;  the vault metrically separable from H. sapiens.

The standard specimen-by-specimen review and the source of the article's table.  Its table is the strongest single piece of evidence for a large range under one name, which is the creationist point.  The same table shows the small values cluster at the base and the large at the top, which is not.  Antón also excludes Kabwe, Bodo, Ceprano, Dali and the Gran Dolina material from erectus, and answers the Langbroek and Roebroeks challenge to the early Java dates on technical grounds.  Obtained through the Wayback id_ route on Wiley's /doi/pdfdirect/ path.

Brain size and encephalization in early to Mid-Pleistocene Homo

Rightmire, G. Philip.  American Journal of Physical Anthropology 124(2):109–123, 2004.  https://doi.org/10.1002/ajpa.10346

camp: mainstream topic: specimens method: measurement direction: complicates verified: abstract

Key figures: mean endocranial capacity 973 cm³ across 30 H. erectus individuals;  1,206 cm³ across ten Middle Pleistocene H. heidelbergensis crania.

The mainstream compilation that sits beside Peter Line's creationist one.  Rightmire's 973 cm³ for 30 individuals and Line's 934 cc for 48 are close enough that the article's claim of agreement between the two camps' numbers holds.  The samples are not the same set, and Rightmire's predates the 2013 publication of Skull 5 at 546 cm³.  Obtained during verification from the OpenAlex Crossref deposit after the earlier attempt through Wiley and the Wayback id_ route had failed;  this reverses the previous "could not be obtained" verdict.

Comment on "A Complete Skull from Dmanisi" and the Response

Schwartz, Jeffrey H., Ian Tattersall & Zhang Chi;  and Lordkipanidze, David, et al.  Science 344(6182):360a and 360b, 2014.  https://doi.org/10.1126/science.1250056, https://doi.org/10.1126/science.1250081

camp: mainstream topic: critique method: commentary direction: complicates verified: abstract

The lumping claim creationists quote as settled was challenged in Science within six months and defended.  The challenge cuts against the creationist reading, because Schwartz and Tattersall want more species, not fewer.  The reply's point, that the critics never calibrated their differences against variation in living species, is exactly the test a "variation within a kind" argument also needs.  Abstracts from the Crossref deposit via OpenAlex.

The Wisdom of the Bones

Walker, Alan & Pat Shipman.  Alfred A. Knopf, 1996.  https://archive.org/details/wisdomofbonesins0000walk

camp: mainstream topic: specimens method: model direction: complicates verified: full

Key figures: Turkana Boy assumed eleven or twelve on human dental standards;  adult stature six feet one inch on a Dinka growth curve;  East African erectus average five feet seven inches and 128 pounds.

Walker's own account of how the famous stature figure was built, and therefore the paper trail for why it fell.  Both assumptions he states openly, a human dental age and a modern adolescent spurt, were later tested and failed.  The substitute for the unobtainable 1993 Nariokotome monograph, by the same first author.  Read by archive.org search-inside;  the leaf-to-page offset drifts through the book and every page cited was read off its own running head.

The dates

Earliest human occupations at Dmanisi (Georgian Caucasus) dated to 1.85–1.78 Ma

Ferring, Reid, et al.  PNAS 108(26):10432–10436, 2011.  https://doi.org/10.1073/pnas.1106638108

camp: mainstream topic: dating method: measurement direction: undercuts verified: full

Key figures: Masavera Basalt ⁴⁰Ar/³⁹Ar 1.85 Ma;  Olduvai/Matuyama reversal 1.78 Ma in the section;  fossils at about 1.77 Ma;  repeated occupation over as much as 80 ka;  73 Mode I artefacts, no bifaces.

The paper that carries the Dmanisi date, and the hardest kind of evidence in the file:  a radiometric floor, a magnetic ceiling, and a biostratigraphic cross-check, agreeing.  Its 600–775 cc figure for the crania predates the CT measurement of Skull 5 and is not used.  Publisher PDF from PMC.

Age of the Earliest Known Hominids in Java, Indonesia

Swisher, Carl C. III, et al.  Science 263(5150):1118–1121, 1994.  https://doi.org/10.1126/science.8108729

camp: mainstream topic: dating method: measurement direction: undercuts verified: abstract

Key figures: argon-argon hornblende plateaus at 1.81 ± 0.04 and 1.66 ± 0.04 Ma.

Fixed the early end of the Java record and supplied the standard explanation for the absence of handaxes in East Asia.  Its other famous Java paper, the 1996 Ngandong dates of 27–53 ka, is superseded and is not cited.  Note the DOI:  10.1126/science.8122351 is not this paper.

Latest Homo erectus of Java: potential contemporaneity with Homo sapiens in southeast Asia

Swisher, Carl C. III et al.  Science 274(5294):1870–1874, 1996.  https://doi.org/10.1126/science.274.5294.1870

camp: mainstream topic: dating method: measurement direction: complicates verified: abstract

Key figures: mean ages of 27 ± 2 to 53.3 ± 4 ka on bovid teeth from the Ngandong and Sambungmacan hominin levels, by ESR and mass-spectrometric uranium-series.

The origin of the 27–53 ka figure that creationists cited for two decades as proof that erectus overlapped modern humans.  Antón 2003 already allowed that a systematic error in the technique could void it;  Rizal 2020 replaced it with 117–108 ka.  Cited in the article only as history, and named so a reader can see exactly where the old number came from.

Age of Zhoukoudian Homo erectus determined with 26Al/10Be burial dating

Shen, Guanjun, Xing Gao, Bin Gao & Darryl E. Granger.  Nature 458(7235):198–200, 2009.  https://doi.org/10.1038/nature07741

camp: mainstream topic: dating method: measurement direction: undercuts verified: abstract

Key figures: weighted mean of six measurements 0.77 ± 0.08 Myr for lower cultural layers 7–10.

The date that matters most for a young-earth argument, because a two-nuclide burial age depends on the difference between two decay constants and does not cancel under a uniform rescaling of decay rates.  No creationist source in the file addresses it.

Homo erectus at Trinil on Java used shells for tool production and engraving

Joordens, Josephine C. A., et al.  Nature 518(7538):228–231, 2015.  https://doi.org/10.1038/nature13962

camp: mainstream topic: behaviour method: measurement direction: supports verified: abstract

Key figures: Trinil bone-layer sediment inside shells dated 0.54 ± 0.10 to 0.43 ± 0.05 Ma by argon-argon and luminescence;  one shell tool;  one geometric engraving.

Supports the creationist reading on behaviour more than any other single find, because engraving is symbolic, and the authors decline to say what it meant.  Also the thinnest item in the file:  one object from a nineteenth-century collection, with the grooves' origin challenged.  The article says both.

Last appearance of Homo erectus at Ngandong, Java, 117,000–108,000 years ago

Rizal, Yan, et al.  Nature 577(7790):381–385, 2020.  https://doi.org/10.1038/s41586-019-1863-2

camp: mainstream topic: dating method: model direction: complicates verified: abstract

Key figures: Bayesian model on 52 radiometric ages by four methods;  bone bed 117–108 ka.

Undercuts the decades-old creationist talking point that erectus at Ngandong overlapped modern humans at 27–53 ka, because the field has abandoned that date.  Supports the creationist observation that the mainstream moved the date by a factor of four in twenty-five years.  Every position in that fight is far outside a young-earth chronology.

Hominins on Flores, Indonesia, by one million years ago

Brumm, Adam, et al.  Nature 464(7289):748–752, 2010.  https://doi.org/10.1038/nature08844

camp: mainstream topic: behaviour method: measurement direction: supports verified: abstract

Key figures: ignimbrite over the Wolo Sege artefacts dated 1.02 ± 0.02 Myr by argon-argon.

Tools on Flores mean a water crossing.  No fossil is associated with them, so the article does not slide from "hominins" to erectus.  The Mata Menge fossils belong to the floresiensis article.

The behaviour

Microstratigraphic evidence of in situ fire in the Acheulean strata of Wonderwerk Cave

Berna, Francesco, et al.  PNAS 109(20):E1215–E1220, 2012.  https://doi.org/10.1073/pnas.1117620109

camp: mainstream topic: behaviour method: measurement direction: supports verified: full

Key figures: burned bone at 400–550 °C by FTIR, 30 m inside the cave;  stratum 10 bracketed by cosmogenic burial ages of 1.27 ± 0.19 and 0.98 ± 0.19 Ma, normal polarity assigned to the Jaramillo subchron.

The strongest early-fire claim because of its method, micromorphology on intact blocks plus infrared microspectroscopy.  The authors say plainly that they found no built hearths.  Undercuts the chronology by the same route as Dmanisi, two independent clocks agreeing.  Publisher PDF from PMC.

Evidence of Hominin Control of Fire at Gesher Benot Ya'aqov, Israel

Goren-Inbar, Naama, et al.  Science 304(5671):725–727, 2004.  https://doi.org/10.1126/science.1095443

camp: mainstream topic: behaviour method: measurement direction: supports verified: abstract

Burned seeds, wood and flint at about 790 ka, with burned microflint clustering in spots read as hearths.  The abstract says "suggestive" and "possibly," not "demonstrates."  The Crossref deposit carries a typo, "live" for "olive," which the article does not reproduce.

On the earliest evidence for habitual use of fire in Europe

Roebroeks, Wil & Paola Villa.  PNAS 108(13):5209–5214, 2011.  https://doi.org/10.1073/pnas.1018116108

camp: mainstream topic: behaviour method: review direction: undercuts verified: abstract

Key figures: habitual fire becomes "a significant part of the hominin technological repertoire" only from about 300 to 400 ka.

The standing mainstream objection to reading Wonderwerk and Gesher Benot Ya'aqov as controlled fire, and Berna's own paper answers it by name.  Its scope is the European record, so neither of the two early sites falls inside it;  that limit is why the dispute is live rather than settled.  Obtained during verification;  previously listed as considered-but-not-obtained.

Earliest known hominin activity in the Philippines by 709 thousand years ago

Ingicco, Thomas, et al.  Nature 557(7704):233–237, 2018.  https://doi.org/10.1038/s41586-018-0072-8

camp: mainstream topic: behaviour method: measurement direction: supports verified: full

Key figures: 57 tools with a 75 %-complete rhinoceros skeleton;  cut marks on thirteen bones;  ESR/U-series on enamel 709 ± 68 ka;  bone bed bracketed 777–631 ka.

The cleanest sea-crossing case in the file.  The authors do not name erectus and do not say boats, and the article keeps to what they claim.  Read in the accepted manuscript at HAL, behind an Anubis proof-of-work wall cleared with bin/anubis-fetch.py.

The earliest toothless hominin skull

Lordkipanidze, David, et al.  Nature 434:717–718, 2005.  https://doi.org/10.1038/434717b

camp: mainstream topic: behaviour method: measurement direction: supports verified: abstract

D3444 with D3900 had lost all but one tooth several years before death.  The 2005 paper says "raises questions," not "proves compassion," and offers soft foods without help as an alternative.  The 2024 paper by the same team makes the provisioning inference in print, which is the stronger way to make the creationist point.

The growth schedule

Growth processes in teeth distinguish modern humans from Homo erectus and earlier hominins

Dean, Christopher, et al.  Nature 414(6864):628–631, 2001.  https://doi.org/10.1038/35102204

camp: mainstream topic: development method: measurement direction: undercuts verified: abstract

Enamel formation rates from daily incremental markings in thirteen fossil hominins, including Nariokotome and Sangiran S7-37.  Neither australopiths nor early Homo show the slow modern human trajectory.  The twenty-three-year-old precursor of Zollikofer 2024, by a different method on different specimens.  Abstract obtained twice by independent routes and identical.  Lubenow's 2004 revision does not engage it.

Just how strapping was KNM-WT 15000?

Graves, Ronda R., et al.  Journal of Human Evolution 59(5):542–554, 2010.  https://doi.org/10.1016/j.jhevol.2010.06.007

camp: mainstream topic: development method: model direction: complicates verified: abstract

Key figures: growth complete by 12.3 years;  adult stature 163 cm, not 185 cm.

Withdraws the six-foot Turkana Boy that Lubenow faithfully reported from Walker.  The creationist figure was not invented;  it was the mainstream figure of the 1990s, and it is now obsolete.

Dental evidence for extended growth in early Homo from Dmanisi

Zollikofer, Christoph P. E., et al.  Nature 635:906–911, 2024.  https://doi.org/10.1038/s41586-024-08205-2

camp: mainstream topic: development method: measurement direction: undercuts verified: full

Key figures: synchrotron microtomography to 0.72 µm;  death at 11.36 ± 0.65 years;  dental maturity 12.0–13.5 years against 18–22 in modern humans;  molar eruption 3.8, 7.6 and 10.6 years, matching wild chimpanzees;  pattern of maturation human-like, rate ape-like.

The strongest single piece of evidence against "erectus is simply a human population," and the newest.  It also hands the creationist side its best point in the same paper:  the excavators infer alloparental provisioning from the toothless elder.  Writers on both sides have quoted one half of the pattern-versus-rate finding and dropped the other;  the article states both.  Open access, CC BY-NC-ND.  No creationist response was found.

The young-earth readings

Bones of Contention, revised edition

Lubenow, Marvin L.  Baker Books, 2004.  https://archive.org/details/bonesofcontentio0000lube_s0n6

camp: creationist topic: yec-argument method: review direction: supports verified: full

Key claims: erectus and Neandertal are one continuum inside Homo sapiens;  modern cranial capacity 700–2,200 cc;  erectus 780–1,225 cc;  post-Flood Ice Age people spreading from the Near East.

The standard young-earth treatment.  Two of its numbers fail on its own sources:  the 780 cc floor omits D2700 at about 600 cc, which is in his endnotes and on his chart, and the 700 cc modern floor rests on three textbook statements of which two disagree with it.  The Trinkaus "Virtually none" exchange is a personal recollection and is held in quarantine.  Read by archive.org search-inside;  page = leaf − 2, anchored twice.

Baraminological Analysis Places Homo habilis, Homo rudolfensis, and Australopithecus sediba in the Human Holobaramin

Wood, Todd Charles.  Answers Research Journal 3:71–90, 2010.  https://answersresearchjournal.org/human-holobaramin/

camp: creationist topic: yec-argument method: measurement direction: complicates verified: full

The best statement of where erectus sits in creationist thinking and of what a post-Babel model owes.  Its survey shows the field split three ways on erectus, and its stratigraphic-order sentence is the most useful creationist quotation in the file.  Its method and the Reeves 2021 critique are handled in the pre-human-hominins article.

Explaining robust humans, and New Dmanisi skull threatens to bring the house down

Line, Peter.  Journal of Creation 27(3):64–71, 2013;  and creation.com, updated 29 October 2013.  https://dl0.creation.com/articles/p091/c09176/j27_3_64-71.pdf, https://creation.com/dmanisi

Keys:  line-2013 is the journal paper;  line-2013b is the web article, the source of the quoted "iodine" and "less likely" passages.

camp: creationist topic: yec-argument method: hypothesis direction: complicates verified: full

The one worked creationist model for the small Dmanisi brains:  robusticity from longevity-linked thyroid patterns, and the Dmanisi brains as cretinism in an iodine-poor region.  Line calls it speculative and judges normal variation, Lubenow's whole argument, less likely.  Cretinism delays dental development, and the 2024 synchrotron result shows fast development, so the model predicts the opposite of the measurement.  His citation practice, checked, is accurate.

Paleoanthropology in Australia—Homo erectus and modern human origins

Line, Peter.  Journal of Creation 32(3):88–96, 2018.  https://dl0.creation.com/articles/p124/c12415/j32_3_88-96.pdf

camp: creationist topic: critique method: review direction: complicates verified: full

A model of how to handle a contested claim:  Line reports Durband against Curnoe on Kow Swamp deformation and notes the collection is gone.  The right citation for the Australian robust crania, in place of the ICR piece.

Homo erectus: The Ape Man That Wasn't

Tomkins, Jeffrey P.  ICR Acts & Facts, 30 September 2019.  https://www.icr.org/article/11538

camp: creationist topic: yec-argument method: commentary direction: complicates verified: full

Better sourced than most popular pieces, and the site of the clearest citation-against-itself in the file:  the Salkhit date is taken from Devièse 2019 while the archaic description that paper overturned is kept.  Its overlap-in-time argument against descent is not used by the article.

Compound-specific radiocarbon dating and mitochondrial DNA analysis of the Pleistocene hominin from Salkhit Mongolia

Devièse, Thibaut, et al.  Nature Communications 10:274, 2019.  https://doi.org/10.1038/s41467-018-08018-8

camp: mainstream topic: critique method: measurement direction: undercuts verified: abstract

Hydroxyproline radiocarbon 34,950–33,900 cal BP and a complete mitochondrial genome in modern human haplogroup N.  Cited to show what Tomkins's own source says.  Open access, CC BY.

New Homo erectus crania associated with stone tools raise issues

Line, Peter.  Journal of Creation 34(2):55–61, 2020;  web version 2021.  https://creation.com/homo-erectus-with-stone-tools

camp: creationist topic: yec-argument method: review direction: complicates verified: full

Key figures: 48 crania, mean about 934 cc, range 546–1,260 cc, about 70 % of the modern mean.

The most honest creationist statement of the problem anywhere in the file.  Line concedes that pathology cannot explain most of the small crania, that the cause is unknown, and that the "devolution from post-Babel moderns" model looks like too much coincidence.  His compiled range agrees with Antón's table.  The printed Journal of Creation PDF was obtained on 6 September 2026 from dl0.creation.com/articles/p137/c13758/j34-2_55-61.pdf, and every quotation in the article now carries a folio between 55 and 61.  Line's note 13 credits the 48-cranium compilation to his own chapter 14 of The Apeman Controversy (2020), so it is his count, not a borrowed one.

Races, Types, and Ethnic Groups (1975) and Human Variation (1992)

Molnar, Stephen.  Prentice-Hall, Englewood Cliffs, 1975 and 1992.  https://archive.org/details/racestypesethnic0000moln and https://archive.org/details/humanvariationra0000moln

camp: mainstream topic: yec-argument method: review direction: undercuts verified: full

Key figures: 1975, p. 57, "in fact, there are many persons with 700 to 800 cubic centimeters," with within-population variation given as plus or minus 400 cc.  1992, p. 148, the sentence is gone and the variation is given as ±100 cc.

The textbook behind the creationist "modern humans go down to 700 cc" floor, reached at the primary during verification after the article had recorded it as unobtainable.  The quotation is real and Lubenow reports it correctly.  It also carries no data in either edition, and Molnar deleted it from the later one while cutting his own spread figure by a factor of four.  This is the single strongest new item found in the verification pass, and it cuts against the young-earth side.

The Pygmy Human Hypothesis

Rupe, Christopher.  Creation Research Society Quarterly 62:208–227, Winter 2026.

camp: creationist topic: yec-argument method: review direction: complicates verified: full

Key figures: p. 223, "small cranial capacity" listed among traits that "should not be considered diagnostic."

Cited here only for one sentence.  Rupe's own argument is that a small brain does not make a fossil an ape, which is aimed at the australopiths;  the consequence for this article is that the co-author of Contested Bones has since dropped the criterion his 2017 book's 700-to-1,400 cm³ range depended on.  The rest of the paper belongs to Lucy and the Pre-Human Hominins.

Contested Bones, chapter 4

Rupe, Christopher & John Sanford.  FMS Publications, 2017.  https://archive.org/details/contestedbones0000rupe

camp: creationist topic: yec-argument method: review direction: complicates verified: full

Sides with the lumpers, then proposes that Skull 5 may be an australopith misfiled as erectus, which is a fourth creationist position on one skull.  Its erectus range of 700–1,400 cm³ is wrong at both ends by its own endnote 15.  Its Dmanisi reporting is otherwise accurate.  Read by archive.org search-inside, which works on this printdisabled item with plain login cookies;  page = leaf − 2, anchored twice.

Considered and not used

• Semaw et al. 2020, the Gona crania. Reached only through Line 2020. The ~598 cc figure for DAN5/P1 is reported as Line's.
• Langbroek and Roebroeks 2000. Reached only through Antón 2003, which quotes and answers it.
• Vekua et al. 2002, Science 297:85–89, the D2700 description. Reached through Antón's Table 3 and Lubenow's own endnote; cited only to show that Lubenow had it.
• Hammond et al. 2021 on KNM-ER 2598, already in the library. Bears on the 1.78 versus 1.65 Ma dispute Ferring notes; not needed here.
• Lubenow's Trinkaus anecdote ("Virtually none"). A personal recollection of a lecture question period. Held in quarantine.
• Rupe and Sanford's quotation of Lordkipanidze ("If you found the Dmanisi skulls at isolated sites in Africa ..."). Second-hand and not checked against the primary. Held in quarantine.

Could not be obtained

• Walker, Alan and Richard Leakey (eds.), The Nariokotome Homo erectus Skeleton, Springer, 1993. The primary monograph on KNM-WT 15000 and the source of the 909 cc figure. Not on archive.org; paywalled at Springer; no repository copy. Substituted by Walker and Shipman 1996 for the stature reasoning and by Antón 2003 Table 3, which cites it, for the volume.
• Rightmire, Lordkipanidze and Vekua 2006, Journal of Human Evolution 50:115–141, the full description of the first four Dmanisi crania. Elsevier closed, no open copy. Its conclusions are reached through Lordkipanidze 2013.
• Full texts of the thirteen abstract-only papers listed above. Publisher sites, Unpaywall, OpenAlex, Semantic Scholar, the Wayback id_ route and three institutional repositories were tried; none supplied a file. An institutional login or document delivery would open them.
• Schultz, one of the three primaries behind the "modern humans go down to 700 cc" claim. Cited only through Lubenow. Molnar, the second, was obtained on 6 September 2026 and has his own entry above.
• Any creationist response to Zollikofer 2024 or Dean 2001. creation.com, icr.org and Answers Research Journal were searched. Nothing was found. Say "none was found," not "none exists."