Article Catalogue

503 of 592 articles written. Drafts by AI are unchecked until they are rewritten by hand, so treat every claim and citation in them as unverified. Dimmed cards are planned articles that do not exist yet. Each heading counts its articles: (3) when all three exist, (2/3) while one is still planned.

Biology (201)

Limits of Evolutionary Change Creationists teach that species change within created kinds, not that species are fixed. Hybrids give real one-way evidence that groups belong together, often across genera and families. The tools for proving a boundary are weak, and no kind boundary has been shown to be an absolute limit on change. Low-Effort Arguments in the Evolution Debate Published Microevolution and macroevolution are legitimate scientific terms; appeals to consensus, credentials, or bad design do not answer the underlying evidence. Semantic and genetic-fallacy arguments should be separated from substantive tests of evolution and design.

Biological Design (54)

Taken together, the six articles find no test yet that design passes and evolution fails. The design inference, irreducible complexity and redundancy each wait on a number neither side has produced, and simple design reuse fails. The flaws critics cite mostly argue common descent, not against design. Parasites raise a moral question that well-made machinery sharpens.

Convergent Evolution and Design Reuse Camera eyes, sonar and wolf-like skulls each arose more than once, but from different parts. Where the same amino acid changes recur, the silent DNA sites still follow the family tree. That fits separate solutions under selection better than one reused design. Appeals to convergence in anatomy and fossils still lack an independent test. Genetic Redundancy Most genes can be knocked out without visible harm, yet nearly all matter under some condition. Measured backup comes from duplicate genes and from unrelated network routes, with duplicates the larger share in yeast metabolism. Evolutionary models explain how redundancy persists but assume the step that gives unrelated genes overlapping jobs. Conservation tests miss only perfect backups. Organisms Appear Designed Published Biologists across viewpoints agree that organisms display intricate, machine-like organization. The dispute is whether natural selection explains that appearance or it warrants design; apparent design is evidence that needs explanation, not by itself proof of a designer. Perspectives on Evolution and Design Being rewritten Scientists across the origins debate agree that data outranks authority. Their disagreements concern how biological complexity arose and whether apparent design is real, while quotations alone cannot decide the evidence. The Design Inference Dembski's explanatory filter removes only the chance hypotheses fed to it. Its one worked biological number prices random assembly, which no biologist proposes, and its author set it aside. A specification may be found after the fact. The test that survives uses a chance model with selection; it has been computed for a few mutations, never for a molecular machine.

Arguments from Bad Design (35)

Biological defects must be judged against measured functions, feasible alternatives, and real constraints. Retinal light guides establish useful function, while appendix research supports a testable protective role. Useful structures can retain costs and inherited limitations. Neither imperfection nor usefulness alone decides whether life was designed.

Atavisms Atavisms are real: extra-toed horses, whale hind limbs, dog hind dewclaws and chicken embryo teeth reawaken retained developmental programs. The human cases fail: true tails lack vertebrae, extra nipples are common embryonic leftovers, horns are skin lesions, and Uner Tan gait is not ape-like. Atavisms show loss within a lineage, not how far back it goes. Bad Design in Biochemistry Fazale Rana answers five biochemical bad-design charges with second uses. The futile-cycle and RuBisCO answers hold. Bilirubin is a real antioxidant whose health payoff is unproven. The immune system uses discarded new proteins but does not explain how many are discarded. Uric acid is a broken gene, and its antioxidant defence failed its tests. Bird Lungs and Mammal Lungs Bird lungs extract oxygen better than mammal lungs, and it matters at altitude. That does not show mammal lungs are badly designed. At sea level the mammal lung is not the bottleneck, bats fly on it, and its collapse protects deep divers. Talking, coughing and panting are not measured reasons: birds do all three. Body Hair, Nails, and Wrinkled Fingers Human body hair and nails have measured uses: fine hair helps people notice crawling insects, follicles speed wound healing, and nails serve touch and strong grips. Water-wrinkled fingertips are nerve-driven, but whether they improve wet grip is unsettled, with two studies for and one against. None of the three separates common design from common descent. Male Nipples Male nipples are a side effect of a mammary rudiment laid down in both sexes before fetal testosterone. Mice, rats, most horses and all marsupials show that males can go without, and male echidnas keep mammary glands without nipples. Men gain some arousal and pay a small cancer cost. The case barely separates design from descent. The Inverted Retina The vertebrate retina combines real optical costs with effective light guidance and photoreceptor support. Müller-cell experiments refute a simple obstruction argument, but do not prove inversion necessary or globally optimal. The evidence supports a functional, constrained system; its geometry alone does not settle design or ancestry. The Mitochondrial Genome Human mitochondria keep 13 protein genes in a small, many-copied genome that mutates about 30 times faster than nuclear DNA, mostly through copying errors rather than oxygen damage. Experiments show a real targeting barrier to moving the genes. The disease cost is real, and plants show the fast rate is not unavoidable. The Recurrent Laryngeal Nerve The recurrent laryngeal nerve performs essential work, but that does not establish that its entire detour is necessary. Development and anatomical variation explain its course and support a historical interpretation. Functional responses correct oversimplified criticism without demonstrating optimal design or refuting common ancestry. Why Eyes Use the Light They Use Eyes use roughly 380 to 750 nanometres because retinal pigments need photons of about 1.7 to 3.3 electronvolts, and longer-wave pigments fire spontaneously at body temperature. Near-ultraviolet costs sharpness and retinal damage but does not ionise. Animals seeing beyond the band pay for it. The band is a shared constraint, not a design flaw, and not proof of design.

Embryonic Leftovers (4)

Recapitulation, the claim that embryos replay adult ancestors, was abandoned by biologists by 1940, so Haeckel's faked plates settle nothing. Vertebrate embryos do share a mid-development stage, measured in gene expression, and common descent explains it naturally. Creationist replies mostly answer the dead claim; the open question is why development runs through that stage.

Haeckel's Drawings and Recapitulation Haeckel printed one woodcut as three species in 1868 and idealised his 1874 embryo plates, deleting limb buds, and textbooks reprinted them for a century. No Jena court ever convicted him, and historians dispute his intent. The fraud does not settle the embryology: von Baer's divergence and the mid-development likeness are measured without his drawings. Pharyngeal Arches and 'Gill Slits' Human embryos never have gills or open slits, so 'gill slits' is a misnomer both sides should drop. The pharyngeal arches are still the same series on which fish grow gills, patterned by the same genes. Their embryonic work is real, but it does not explain why the parathyroid shares a gill-bud gene. The Embryonic Tail, Lanugo, and the Pronephros The human embryo builds a segmented tail and then kills about five pairs of its segments, the strongest sign of ancestry here. Lanugo's one proposed job, holding vernix, has never been tested. A human pronephros probably never forms; the mesonephros does, and its duct is needed for the permanent kidney and male ducts.

The Human Body Under Criticism (6)

Critics from Krogman in 1951 to Lents in 2018 use human anatomy mainly as evidence of history. The one whole-body redesign concedes its own costs and has never been priced. Back pain and hernia are large costs; choking and the nuchal cord are small. The Fall answer covers wear and disease, not layout.

Childbirth and the Obstetrical Dilemma The human newborn's head fits the birth canal with millimetres to spare, and without care obstructed labour kills. But historical maternal deaths were mostly infection, toxaemia and bleeding, and the walking cost of a wider pelvis is near zero. The real trade-off is with the pelvic floor, which a designer would face too. One Passage for Food and Air Every mammal's food path crosses its airway; adult humans add a low larynx. Choking kills about 5,900 Americans a year, three quarters of them over 65 and mostly with dementia, stroke, missing teeth or alcohol. Speech clarity may need the low larynx; speech itself did not. Every separated route built so far costs voice, mouth breathing or smell. The Spine and the Cost of Standing Up Back pain and knee arthritis are large, real costs, and wild apes show far less spinal disease than people. But knee arthritis doubled after 1950 independent of age and weight, disc failure is mostly inherited, and the only drawn redesign of the back would raise disc pressure. The cost is real; no better build has been shown. The Umbilical Cord No named scholar was found calling the umbilical cord bad design. A cord loops the neck or body in a fifth to a third of births and rarely harms. Short and long cords both raise perinatal death, so length is a trade-off. In horses, cord problems are the commonest diagnosed cause of lost foals. The Urogenital Layout The vas deferens loop has no known medical cost, as George Williams, who raised it, admitted. The prostate around the urethra costs much late in life, but no rebuilt layout has been priced. The testes' route through the groin is the real cost: men's lifetime hernia risk is about nine times women's.

Vestigial Structures (15)

The old vestige argument, that useless organs prove evolution, is dead, and most listed organs do work. The living argument is about reduced homologues and ancestry, which finding a function does not answer. Whale and snake pelves and horse splints remain strong for ancestry; the appendix, goosebumps and palmaris longus are weak for either side.

Blind Cave Fish The Mexican tetra's cave eye forms in the embryo and is then dismantled by lens death under expanded hedgehog signalling; twelve eye loci, oca2 and cbsa lesions and inter-cave crosses show real polygenic loss. Creationists rightly call it loss, but every nuclear clock puts the caves at 20,000 to 190,000 years, which no creationist source has priced. Dewclaws and Carpal Pads The front dewclaw of dogs and cats is a shortened working thumb with its own muscles, not a vestige. The real hind-limb remnant is a tiny first metatarsal, which does fit the vestige test; a dog's hind dewclaw is a toe regained by mutation. Nobody has measured what the carpal pad does, or argued that it is a vestige. Ear Muscles The human ear muscles are the facial-nerve muscles that swivel the pinna in cats, horses and monkeys, reduced to millimetres of movement, yet the brain still drives them toward attended sounds. They pass the homology and reduction steps of the vestige test; the ancestral step rests on the primate tree. Creationist replies answer only the useless claim. Flightless Bird Wings The wings of ostriches, kiwis, the Galapagos cormorant, rails and steamer ducks are reduced flying wings, each flightless group nested among flying relatives, and creationist authors agree they descend from fliers. The case passes all three steps of the vestige test. It proves fast, repeated loss of flight, which both sides predict, and nothing beyond the family. Goosebumps Goosebumps are the arrector pili muscles pulling vellus hairs upright. The muscle is not vestigial: it anchors the nerve that runs hair-follicle stem cells, and its loss tracks baldness. The reflex is intact and homologous with fur-raising in other mammals. What is reduced is the hair, and whether that reduction is ancestral is a question goosebumps cannot answer. Horse Splint Bones The horse's splint bones are reduced second and fourth metacarpals and metatarsals. They bear the small carpal bones, house the suspensory ligament and often fracture, so are not useless. Homology, five embryonic digit condensations and extra-toed horses make them the strongest reduced-homologue case in a living animal; creationist replies answer only the useless claim. Palmaris Longus and Plantaris Neither muscle is useless: palmaris longus flexes the wrist and helps the thumb, and plantaris is a weak calf flexor. Among primates only gorillas, chimpanzees and humans often lack the palmaris, which fits common descent. The plantaris lost its link to the sole in the ancestor of Old World monkeys and apes, so it says little about humans. Snake Pelvic Spurs Boa and python spurs are working hind limbs in miniature: a femur under a horny claw, on a pelvic rod that no longer meets the spine, used in courtship and combat. Python embryos start a leg and stop, every snake sequenced shares one 17-base deletion in the limb enhancer, and Cretaceous snakes had sacra and legs. Creationists concede the reduction. The Appendix The human appendix is a working lymphoid organ, not a useless stump. Darwin's shrunken-caecum story fails across mammals, where the organ arose 29 to 41 times and was rarely lost, though apes pair an appendix with a small caecum. The safe-house role is plausible but unproven, appendicitis is a diet-linked cost, and the organ belongs on neither side's vestige list. The Coccyx The human coccyx anchors the pelvic floor and bears weight in sitting; Darwin claimed only that it is functionless as a tail. Its three to five arch-less centra, its optional tail-extensor muscle and the ape-only Alu insertion in TBXT make it the cleanest reduced homologue in the human body, so the argument turns on ancestry, not use. The Panda's Thumb Gould's panda's thumb fails as a bad-design argument: he measured no shortfall, the grip is fast and precise, and a weight-bearing tradeoff plausibly holds the bone short. It partly survives as an argument from history, since the thumb is a remodelled wrist bone with a useless twin in the foot, which young-earth creationists now place within the bear kind. The Plica Semilunaris The plica semilunaris is a small fold of conjunctiva that steers debris out of the eye and carries immune tissue, so it is not useless. Its match with the third eyelid of other mammals is well supported. It is reduced against lemurs and lorises, but whether that primate pattern reflects ancestry or need has not been tested. Whale Pelvic Bones Living whales carry a pair of small bones, detached from the spine, that anchor the penis muscles in males and the genital muscles in females; a 2014 study shows their size tracks testis mass. The bones are functional and reduced at once. Creationists answer the useless-organ claim but mostly not the homology claim, which rests on embryos, atavisms and fossils. Wisdom Teeth and the Modern Jaw Third molars are smaller and more often absent than the molars ahead of them, and on the mainstream fossil series they have shrunk since Homo erectus. Today's impaction epidemic is developmental: jaws raised on soft food grow shorter, as Darwin proposed and Corruccini, Lieberman and von Cramon-Taubadel confirmed. Diet answers the bad-design charge; ancestry remains a question about the tree.

Irreducible Complexity (3)

Behe's 1996 definition rules out only direct routes, as he said on his page 40. Whether indirect routes are too improbable has never been calculated for any named machine. Muller's 1918 route was shown in a laboratory in 2012, for a duplicated part that lost function. No route to any of Behe's five systems has been reconstructed with measured fitness.

Blood Clotting Behe limited his clotting claim to four factors past the pathway's fork. Knockout mice and every vertebrate genome examined support it, though factor V in jawless fish rests on gene structure alone. Doolittle's 1997 mouse counterexample was a misreading, and the whale and puffer-fish losses fall before the fork. No step building the core has been measured. The Bacterial Flagellum The flagellum's forty parts are genes from knockout screens. About twenty proteins form the structure, and six can be deleted. The type III secretion system is its descendant. About eleven of 42 proteins have outside relatives. The oldest reconstructable flagellum was already complete, and no stepwise model has a measured step.

Predation, Parasites, and Pathogens (10)

Gene loss, altered regulation, and reused biological machinery can help explain how harmless relationships become harmful. These mechanisms support parts of creationist corruption models, but do not establish an originally harmless world or explain every specialized apparatus for attacking a host.

Bacterial Pathogens as Broken Bacteria Gene loss and failed regulation help explain some bacterial diseases, but reduced genomes also support beneficial partnerships. Comparative and experimental evidence supports particular breakdowns, not universal ancestral harmlessness or a date for the origin of disease. Behaviour-Manipulating Parasites The zombie ant's timed death grip and the castrated crab's brood care are well measured. Toxoplasma's effect on rodents is real but not cat-specific, human personality claims have not held up, and birds eating snail broodsacs is untested. Young-earth writers name no original function for any of the five parasites. Brood Parasites and Cheating Species Cuckoos, cowbirds, honeyguides, cuckoo bees and slave-making ants carry working equipment for exploiting another species' parental care. Measurable losses exist, in slave-making ant workers and their smell and taste genes, but they follow the parasitic habit. Young-earth writers place the origin after the Fall and admit the route is unknown. Parasitic Plants and Fungi Parasitic plants shed photosynthesis genes in a measured, repeatable order, and the Rafflesia family has no detectable plastid genome. The young-earth loss model is confirmed gene by gene, but loss follows dependence rather than causing it. The haustorium, host-derived genes and the frog chytrid's protease expansions are gains that no loss model supplies. Parasitoid Wasps Some parasitoid wasps combine inherited viral machinery, changing venom chemistry, and specialized host relationships. These findings support biological reuse and viral ancestry, but do not establish originally harmless relationships. Creationist proposals must explain particular life cycles rather than treating every parasitoid as a product of simple deterioration. Predation, Venom, and Carnivory Cat claws, pit-viper heat pits and fangs, and the cone snail's insulin harpoon are working weapons, not decayed parts. Young-earth writers show a tested pre-Fall use only for claws and sharp teeth. What holds: plant-eating carnivores, venom built from ordinary proteins, predators taking the sick, and kills that take minutes. Ticks, Fleas, Lice, Mites, and Botflies Lice and vampire moths have an evidenced harmless prior diet, and face mites and feather mites are mostly harmless now. Fleas and ticks have only proposals. The tick's saliva, with protein families found nowhere else, and the botfly, whose every species is an obligate parasite, are where the young-earth answer is weakest. Viruses and Phages Most of the roughly 10^31 virus particles on earth infect microbes, and measured benefits to hosts are real. The pathogenic viruses whose origins are dated arose by host jumps and gene capture, as young-earth writers expect. But phages also carry the cholera toxin, most viral genes match no host gene, and the escaped-gene model remains an untested hypothesis. Worms in the Human Body Hookworm and the large roundworm have harmless relatives, so a young-earth account in which they degenerated from harmless ancestors fits the evidence. Whipworm, the filarial worms and Guinea worm have no close harmless relative, and the one published proposal for river blindness conflicts with the nematode tree. Pig-whipworm therapy failed its two large trials.

Evolutionary Mechanisms (16)

What mutation, selection, drift, recombination, and shared patterns can establish about biological change. This page is the front door to the section and links to all of it; the summary and verdict are still to be written.

Development and Constraint Breaking an early patterning gene usually kills, but switch changes beside the same genes alter one body part and have reshaped colour, hairs, spines and teeth between species. None has built a new organ or body plan. Epigenetic marks fade within generations, kin selection explains animal cooperation, and the extended synthesis is a dispute of emphasis. Evolutionary Computation: What Genetic Algorithms Show About Evolution Genetic algorithms work, and every famous one works because a designer drew the search space, wrote the score, and smoothed the landscape until selection could climb it. Avida, ev, Weasel and Tierra run at mutation rates, selection strengths and fitness effects with no biological parallel; the one honest simulation, Tierra, plateaued. The critics of these programs overreach too, and the whole dispute reduces to one unmeasured number: how smooth the real fitness landscape is.

Biogeography and Common Descent (3)

Where animals live tracks sea depth, island order and the nearest source, as descent from local ancestors predicts. The post-Flood model claims these patterns too, through diversification within kinds, and its rafting holds up. It fails on the record: no walkable route to Australia, no marsupial fossils in Asia after the Miocene, and a one-way faunal break.

Creationist Biogeography: The Post-Flood Dispersal Model Rafting, land bridges and rapid speciation are all real, and the model built from them still fails. Measured here: the shallowest walkable route from Asia to Australia needs a 1,264 m sea-level fall, and the Asian fossil trail marsupials should have left is empty. Evolutionary Biogeography: What the Map of Life Shows Darwin predicted in 1859 that the depth of the sea between an island and the mainland would predict its mammals. Measured here on modern bathymetry, it does, across nineteen crossings. The vicariance half of the modern theory has fared much worse, and the field says so itself.

Molecular Clocks (2)

Molecular-clock dates are a ratio with an imported time. Fossil-calibrated deep dates borrow the radiometric timescale and lean old, yet fossil-free family rates give long dates too. Clocks pass tests inside their sampling window and fail toward youth beyond it. Young-earth mitochondrial clocks fit only by counting mutations that do not last.

Do Clock Dates Agree with the First Fossils? No. Clock dates for the great radiations run older than the first fossils, by about 13 to over 100 million years on the mainstream timescale. The gaps have narrowed, but calibration and rate-model choices move them as much. Fossil order still matches molecular branching order. The dispute tests two methods, not the earth's age.

Phylogenetic Inference and Congruence (2)

Biological trees agree above chance, but designed systems can too. Common descent gains force from genomic insertions, chromosome order, and fossils, not congruence alone.

Shared Genetic Mistakes and Common Ancestry Shared broken genes and mobile-element insertions are the strongest genetic evidence for common ancestry. Homoplasy in these markers is real and documented, but measured at 0.05% to 3%, far too rare to explain the pattern. What the failures do break is the claim that the markers are perfect and need no statistics.

Population Genetics (6)

Effective size is not headcount; most mutations are lost by drift; selection weakens below 1/(2Ne); and load is not a death count. Clock and bottleneck estimates depend on demographic models.

Beneficial and Harmful Mutations Measured new mutations with a detectable effect are mostly harmful: 20-41% of random point mutations in small viruses are lethal, and 0-6% of those with an effect in significance-tested viral sets are beneficial. Sanford's one-in-a-million is one 1998 model estimate whose assumptions its first author later revised; other methods give beneficial rates about a thousand times higher. Genetic Entropy Load does not imply extinction, but weak harmful mutations can fix by drift and Muller's ratchet is real. Long-term fitness loss remains unmeasured. Haldane's Dilemma and the Waiting-Time Problem Haldane's reproductive limit is real. On its corrected form, the roughly 4,000 measured adaptive protein changes in the human line fit even as sweeps from new mutations; tens of thousands of non-coding ones would strain it. Waiting-time claims assume prespecified sequence targets, a premise not established. Neutral Theory Most DNA differences between mammal species are neutral or nearly so, chiefly because most of the genome shows no constraint, so the theory's genome-wide form leans on that. In proteins of large-population species half or more of changes look adaptive; in humans about a tenth. The 2018 challenge concerns proteins and linked selection, not junk DNA. Population Bottlenecks A brief bottleneck keeps most heterozygosity but destroys allelic richness. Two founders cap a locus at four alleles. Real founder populations, some from a single pair, recover in numbers and mostly stay damaged in genes. The size of the biblical bottleneck is defensible; its date is not.

Human Origins and Genetics (42)

The fossils and genomes agree with Genesis on what humans are: one shallow family, distinct from the apes, making symbols as early as their bones appear. They disagree on when: rock and genetic clocks put human history several to hundreds of times past a biblical date, and the human-ape likeness follows the pattern of common descent.

Behavioral Modernity Engraving, beads, pigment and burial appear with the earliest Homo sapiens and with Neanderthals, not in a late 50,000-year revolution, as young-earth writers expected. But their dates, about 164,000 to 21,000 years by methods that agree, would need radioactive decay after Babel to run 12 to 250 times faster than today. Human Uniqueness: Language, Mind, and FOXP2 No ape has shown grammar. Apes learn a hundred or more symbols, Nim's signed utterances stayed 1.1 to 1.6 signs long, and Kanzi understood requests about as well as a two-year-old. FOXP2 is a speech-motor gene, not a language gene, and Neanderthals shared its human form. The gap shows human uniqueness, not how it arose.

Hominin Fossils (24)

The young-earth division of the hominin fossils into apes and humans holds at both ends: the australopiths were apes that walked upright, and the Neanderthals, Denisovans and Middle Pleistocene people were human. It splits on the small-brained mosaic fossils between, and has not explained the dates or why brain size rises with time.

Ape or Human? How Creationists Sort the Fossils Young-earth writers agree that the australopiths were apes and that erectus, Neanderthals and the Middle Pleistocene people were human. They split on habilis, rudolfensis, sediba, naledi and the island forms, because each weighs different bones. The split follows mosaic anatomy, not missing data, which is the skeptic's strongest point. Brain Size and the Order of the Fossils Brain size rises with fossil date: in 91 percent of dated skull pairs the younger has the bigger brain, with or without species labels. About half tracks body size. Young-earth writers accept the order but have not explained it. Naledi and floresiensis break its lower edge; only Yunxian 2, newly redated, challenges its upper.

Lucy and the Pre-Human Hominins (6)

Lucy was not human: her brain and locomotor anatomy are australopith, though she walked upright. The oldest biped claims rest on few, reconstructed bones, and mainstream work has reversed the Ardipithecus headline. Homo naledi leans human on both sides' methods. The family trees rest on a method that failed its own controlled test.

Australopithecus sediba Australopithecus sediba was a small-brained, tree-climbing australopith with some Homo-like teeth, hands and hips. It was not human, and it is probably too young to be the ancestor of Homo. Most young-earth writers call it an ape. Todd Wood's skull statistics group it with humans, but his own skeleton statistics do not. Lucy and Australopithecus afarensis Lucy was not a human. Her 388 ml brain, ape-like brain organisation, balance organs and shoulder growth, and young-earth baraminology all say so. Nor was she an ordinary ape: she walked upright. Christopher Rupe's pygmy-human case fails, though he rightly defends the pelvic reconstruction, and the mixed-skeleton claim fails on excavation. Paranthropus The robust australopiths were small-brained, small-bodied bipeds with the biggest cheek teeth of any hominin. Tooth chemistry shows East African boisei ate mostly grasses or sedges, not nuts. Every young-earth writer calls them non-human, but their enamel proteins sit nearer humans than apes, and their dates need post-Flood accelerated decay. Sahelanthropus, Orrorin, and Ardipithecus The three oldest claimed hominins rest on little bone. Sahelanthropus is nine cranial specimens and three limb bones, chimpanzee-like in shape. Orrorin is thirteen fossils. Ardipithecus has over 100 specimens, but its pelvis is a computer reconstruction, and mainstream work since 2019 has substantially reversed its 2009 headline. The South African Australopiths The Taung child, Australopithecus africanus and Little Foot were small-brained bipedal apes, not humans, and young-earth baraminology agrees. Their cave dates disagree by about a million years, between 2 and 3.7 million. A young-earth reading must fit the whole cave history after the Flood and make radioactive decay run hundreds of times faster after it.

The Genus Homo (15)

On what they were, the young-earth reading of Homo as Adam's descendants fits the Neanderthals, Denisovans, Middle Pleistocene people and earliest Homo sapiens well. It is weakest at the small-brained edge, where its own writers disagree. On time it has no answer: every species is dated about 7 to 650 times the years since the Flood.

Denisovans Denisovans are known from a Siberian genome, a Tibetan jaw and a Harbin skull identified by proteins and DNA. They carried the fused human chromosome 2, had large brains and interbred with Neanderthals and modern humans, so reading them as people is well supported. Their dates, 146,000 years and older by uranium decay and genetic clocks, have no young-earth reply. Homo erectus Homo erectus made fire, butchered a rhinoceros across a sea barrier, and kept a toothless elder alive. But its adult brains ran 546 to 1,251 cc, its children reached dental maturity at 12 to 13 years against 18 to 22 today, and seven dating methods agree across three continents. No creationist source answers the teeth or the dates. Homo floresiensis: Island Dwarfing and the Small Brain Island dwarfing is measured and fast, and it explains the hobbit's body. No published model shrinks a modern brain to 426 cc. The models work from a start of about 600 to 860 cc, and best from Homo habilis. Todd Wood's baraminology makes habilis human, an answer no creationist treatment of the hobbit read here has used. Homo floresiensis: The Disease Proposals Four diagnoses of the Flores skeleton LB1 as a sick modern human, microcephaly, Laron syndrome, cretinism and Down syndrome, each fail on a measurement: an ape-pattern wrist formed by the eleventh week and found in two individuals, Lucy's limb proportions, and a brain smaller than cretinism or Down syndrome produces. The critics were right about the skull's volume. Homo floresiensis and Homo luzonensis The Flores and Luzon bones are a real small-bodied population, not diseased modern humans. Island dwarfing explains the body, but published models reach the 426 cc brain from a start of about 600 to 860 cc, never from a modern one. No young-earth source dates either site in years; two place Liang Bua on opposite sides of the Flood. Homo habilis and Homo rudolfensis The earliest Homo fossils have 500 to 800 cc brains and, where a body is known, small bodies with long arms. Mainstream scientists dispute their genus; young-earth writers call them ape, human or a mix and disagree on the same skulls. The KBS Tuff episode shows real bias, not dates wrong by a factor of 400. Homo naledi: Burial, Engravings, and Fire eLife's editors rated the 2023 Homo naledi burial evidence inadequate, with all four reviewers agreeing. In 2025 the two who returned split, the features are still unexcavated, and the stone is still in plaster. The engravings are undated. No paper has reported the fire evidence, and radiocarbon dates on one hearth were young and are unpublished. Homo naledi: The Date Homo naledi's 241 to 335 thousand year date is a model-dependent bracket. Its floor is one recrystallised flowstone, its ceiling a limit the authors call an over-estimate, and its stratigraphy was revised after the dates came in. Every mainstream figure is at least six times any young-earth placement, and no creationist answer addresses the flowstone ages. Homo naledi and the Rising Star Claims More than 1,500 bones of a small-brained hominin were found 30 m underground in a South African cave. The mainstream calls it Homo, creationist baraminology calls it human, and a 2026 enamel proteome found no male and no variation in 20 individuals. The excavators' data count against water, carnivores and a hard fall; how the bodies arrived is unsettled. Neanderthals Neanderthals interbred with modern humans and left descendants in every living population. They buried their dead and built inside caves. Reading them as fully human is well supported. Their dates are not answered: Bruniquel at 176,500 years and a Y-chromosome clock at 370,000 have no young-earth reply, and the one young-earth radiocarbon curve puts the last Neanderthals before Babel. The Middle Pleistocene Muddle A few dozen skulls and one pit of at least 28 bodies, brains about 1,100 to 1,420 cm³, that the mainstream cannot agree to name. Each named ancestor of modern humans is now reconsidered or disputed, which supports, without proving, one variable human kind. The dates do not: independent methods agree, and no young-earth source answers them. The Middle Pleistocene Muddle: The Creationist Literature Lubenow, Line, Rupe and Sanford, Wood, Answers in Genesis and ICR read the Middle Pleistocene fossils as one human kind dispersed after Babel. Their citations hold on the large points and fail at seven places. Their baraminology has tested Kabwe, Petralona, Dali and the Sima DNA, but not Bodo, Arago or Harbin. The Middle Pleistocene Muddle: The Dates Independent methods agree at Mauer and Sima de los Huesos, and where the specialists disagree they do so by 1.1 to 1.6-fold, or tenfold at worst, against the 14 to 200-fold gap between every measured age and a Flood date. The one case of two methods agreeing and being wrong ran too young. No young-earth source names a faulty input. The Oldest Homo sapiens Jebel Irhoud, Florisbad, Omo Kibish, Misliya and Herto are human fossils on any reading. Their dates, about 315,000 to 155,000 years, come from methods that agree where they overlap and stand 28 to 80 times any Flood date. Young-earth writers place them after Babel but name no faulty measurement.

Human Genetic Ancestry (6)

Human DNA fits one origin for all people, and three of four standard arguments against a single first couple fail. It does not fit a recent one: measured rates of lasting change put mitochondrial Eve and Y-chromosome Adam over 100,000 years back, and Africa's genetic depth exceeds any post-Babel dispersal.

Adam, Eve, and Human Genetic Diversity Three of the four standard arguments against a single-couple human origin do not test it, and BioLogos cut its own claim 36-fold without a correction notice. Two problems survive, unmet by any young-earth calculation: HLA-DRB1 introns in five to nine deeply different families where a couple carries four, and a median four-lineage coalescent age of 358,000 to 495,000 years. Archaic Admixture Living people carry Neanderthal DNA in every population studied, and Melanesians carry 4 to 6 percent Denisovan DNA, so the archaic humans had fertile children with modern humans. The archaic ancestry is depleted on the X chromosome and in testis genes, a sign of partial hybrid sterility that no creationist has answered. Mitochondrial Eve and the Young-Earth Timeline Mitochondrial Eve was one woman in a larger population. Family studies find mutations fast because they count new variants, most of which are later lost. Counts of lasting changes in families, ancient genomes and the tree agree, and put her more than 100,000 years ago. A young-earth date needs inherited founder variation or a past mutation burst, neither yet shown. Out of Africa or Out of Babel? Refitting Ramachandran's heterozygosity gradient from the raw public genotypes puts the source of the human dispersal in Africa, not Mesopotamia: R-squared 0.859 from Addis Ababa against 0.647 from Babylon, under every routing convention tried. But drop the African populations and the gradient stops discriminating, so what it fixes is a direction, not a birthplace. Y-Chromosome Adam and Noah The human Y tree has no pedigree-versus-phylogeny gap. Calibrated on the largest direct pedigree measurement alone, with no fossil or radiometric input, its root is 163,000 years old. Its backbone is a ladder of two-way splits, not the three-way split Noah's sons predict. The young-earth clock rests on a filter choice, not on coverage.

Humans and Chimpanzees (6)

Human and chimpanzee DNA is about 96 percent alike in the regions that line up and 85 percent by a whole-genome count. Shared retrovirus positions, nested inversions and mutation's fingerprints fit common descent better than separate creation, and the chromosome 2 fusion is real and old. Young-earth critics corrected real overstatements. Brain, gait, culture and life history differ greatly.

Chromosome 2 Fusion Human chromosome 2 carries a head-to-head telomere array at 2q13 and a dead centromere at 2q21 matching the chimpanzee 2B centromere, both in Neanderthal and Denisovan genomes. Against 298 real fusions the site holds too much telomeric repeat, not too little. The fusion is real, its condition does not date it, and it does not settle common ancestry alone. How Similar Are Human and Chimp DNA? Most human-chimp figures from 84 to 98.8 percent are correct arithmetic on different denominators; 80 percent subtracts the same DNA twice. The 98.8 slogan counted substitutions only. In the 2025 genomes, 11 of the 16 points creationists cite also separate two humans, at least 5 are species difference, and the controlled metrics order the apes as common descent predicts. Shared Retroviruses and Pseudogenes Humans and chimpanzees share about 100,000 endogenous retrovirus positions; the 2005 genome comparison found 361 lineage-specific ones. Measured integration preference predicts 480 to 960 shared positions by chance, two orders short, and closing the gap needs a target pool 100 times smaller than the measured one. Many ERVs and pseudogenes are functional, which does not answer the ancestry question. Synteny, Gene Order, and the Y Chromosome About 89 percent of the human and chimpanzee genomes lie in the same order. The 171 inversions between them sit on duplicated DNA, recur in living people, and mostly nest as the ape tree predicts. Gene order is partly functional, but function does not explain that pattern. The very different Y chromosomes differ as much inside proposed created kinds. The Human-Chimp Split and the Mutation Rate Mutation rates counted in families run half the old fossil-calibrated rate, pushing the human-chimp split from about 4 to between 5.5 and 13 million years. No steady rate fits every calibration. Mutation since creation could make under 0.1 percent of the differences, so young-earth models call them created, yet they carry mutation's fingerprints.

Race, Skin Color, and Common Ancestry (3)

Humans are one shallow species: 85 to 95 percent of genetic variation lies within every population, yet thousands of markers together still sort people by continent. Skin color is many-gened and shifts fast in populations. Polygenism failed. The shallow pattern fits a recent African origin and a young-earth family alike; allele ages and dating separate them.

Scientific Racism and Monogenesis Polygenism, the nineteenth-century science of separately created races, was wrong: genetics shows one shallow, interbreeding human population. Darwin, Haeckel and Galton ranked races and fed eugenics, though Darwin hated slavery. Christians kept one origin, yet many defended slavery from Noah's curse, and early creationists carried racial ideas the movement later rejected. The Genetics of Skin Color Skin colour is polygenic, not a two-gene trait. Four loci explain 29% of African pigmentation variance, and in the KhoeSan a heritability of 0.97 is almost entirely unexplained by known genes. Populations do shift fast from standing variation. A single mid-brown couple does not produce the human range in one generation, and misses by three orders of magnitude.

Junk DNA and Genome Function (9)

How much of a genome does anything, and what follows if most of it does. Evolutionary theory long predicted that large genomes are mostly junk; design advocates predicted widespread function. This is the debate where that prediction is settled or not.

Functional DNA is Evidence of Design Published At least 85% of human DNA is transcribed in cell- and stage-specific patterns, and tested transcripts are often functional. The article argues that most mammalian DNA serves a purpose and that this exceeds what mutation and selection can create or preserve. Functional versus Junk DNA Predictions Published From 1968 through 2014, many evolutionists predicted that most large genomes would be junk, while design advocates predicted widespread function. Later discoveries of pervasive transcription and regulatory activity favored the design prediction, though biochemical activity does not prove that every base is sequence-constrained. GWAS and Noncoding Variants About 96% of disease-linked variants lie outside protein-coding DNA, but per base they are concentrated in coding and regulatory DNA, and most are tags for nearby causes. Proven noncoding causes exist and matter. The data show regulatory DNA is a large share of what matters, not that most of the genome does. Noncoding RNA and Pervasive Transcription Most of the human genome is copied into RNA, but most of that is introns of known genes, most lncRNAs sit at a few copies per cell, and only a minority tested by knockout show an effect. Random DNA is transcribed in yeast but silent in mouse stem cells. Most noncoding RNA is untested, not proven functional or junk. Overlapping Genes and Multi-Layer Coding Two-frame coding fills half the hepatitis B genome but 0.064% of human coding letters on one standard transcript per gene. Where it occurs it couples two proteins: about 1% of single-base changes are silent in both. Where comparisons date them, second frames look lineage-limited, as overprinting predicts. Claims of many codes on most DNA are unmeasured. Pseudogenes About 14,700 human pseudogenes cover under half a percent of the genome. About one in five shows transcription and a few percent show constraint, but only a few dozen have a measured function. Most were never tested. HBBP1 is a real relic-turned-useful case; vitellogenin, enamelin and GULO are not. Synonymous Mutations and Codon Usage Some synonymous mutations change splicing, mRNA stability or folding, and cause disease. Most measured in human cells have no detectable effect, and the 2022 yeast claim that most are strongly harmful fails there. Ape comparisons put 20 to 40 percent of these sites under weak constraint: enough to bias dN/dS and clocks modestly, not to overturn them. The C-Value Paradox Published Large genomes need not be mostly junk: genome size affects cell size and development, while repetitive and noncoding DNA can have structural and regulatory functions. The onion test therefore does not establish that organisms are undesigned. Transposable Elements Transposable elements are 46% of the human genome. A few dozen copies became genes or switches proven by knockout or deletion. About 16% of TE sequence carries a regulatory mark, at or below chance, about 3% is conserved, and almost none has been tested. Neither 'mostly junk' nor 'mostly functional' is measured.

Major Evolutionary Transitions (6)

Across five proposed steps from oxygen-making photosynthesis to animal body plans, the parts are related as common descent predicts, and something short of each transition is seen or built. The step that defines each, from water splitting and the nucleus to meiosis, the germ line and new body plans, is neither demonstrated nor priced.

Endosymbiosis and the Eukaryotic Cell Mitochondria and plastids carry bacterial genes, ribosomes and gene order that common design does not predict, and living cells show capture, gene transfer and protein import. Relocating genes is fast only if selection favoured each move. How an archaeon became a eukaryote with a nucleus has no intermediates and no demonstrated pathway. The Origin of Multicellularity Cells that stick together have evolved in the laboratory within weeks, in yeast by breaking one gene. No experiment has produced specialised cell types or a germ line. Volvox shares almost all its genes with single-celled Chlamydomonas, so its division of labour rests on regulatory changes nobody has watched or measured. The Origin of Nervous Systems and Animal Body Plans Neurons arose either twice or once with later losses, and no route to the first nerve net is shown. The shared Hox toolkit fits common ancestry. Early development and its gene networks do change between living species, against Meyer and Nelson, but no experiment or fossil shows a new body plan being built. The Origin of Oxygenic Photosynthesis Photosystem II is related to the reaction centres of bacteria that make no oxygen, and each photosystem type works alone in some bacteria. But no living intermediate, laboratory route or priced pathway reaches the four-manganese cluster that splits water, and when and how the two photosystems joined is unsettled. The Origin of Sex Sex costs about twice as much as cloning, and snail experiments measured that cost. Parasites, harsh environments and harmful mutations each make sex pay in particular species, but no mechanism has been shown to pay it everywhere. Meiosis was already complete in the earliest eukaryotes, and how it began has no demonstrated pathway.

New Genes (5)

Comparing genomes shows where most genes got their templates: copying, with two to fourteen amino-acid changes for each tested new job, and a few short genes from non-coding DNA. No route reaches the thousands of founding protein families in each genome, and those founders are where the dispute over how rare a working protein bites.

Antifreeze Proteins Antarctic notothenioids and northern cod built near-identical antifreeze genes by separate routes, one from a digestive-enzyme gene and one from non-coding DNA. The evidence is strong but comparative; nobody watched it happen. What arose is a short repeated unit copied hundreds of times: genuinely new genes that carry little new information. Gene Duplication and Divergence Gene copies arise constantly and most are lost within a few million years. Copying can supply the template of about 40 to 76 percent of the genes in human, yeast and Arabidopsis, and measured new functions took 2 to 14 amino-acid changes. It cannot explain the 1,800 to 7,300 founding families in each genome. How Rare Is a Working Protein? Two experiments measured how rare a working protein is and disagreed by sixty-six orders of magnitude. About half the gap is length; the rest is a harder target and two unmeasured terms. An independent estimate agrees that any one large fold is astronomically rare. No probability means anything without the number of trials. Hoyle's 747 and the second-law argument fail. Orphan Genes Orphan status means no detectable relative, not a demonstrated new gene. Reanalyses reduce several published counts, but annotation changes alone do not prove nonfunction. Verified de novo examples exist; their frequency, molecular complexity and ancestry require evidence beyond a failed similarity search.

Observed Evolution (26)

What evolution has been watched doing, and what it produced. Every case here rests on change seen happening rather than on comparative genomics across taxa; origin questions settled by sequence comparison sit under New Genes instead.

Benchmarking Evolution (9)

A ledger of births, mutations and adaptive changes for HIV, influenza, SARS-CoV-2, tuberculosis, the malaria parasite, Lenski's bacteria and phage lambda. One-step changes are cheap everywhere. A change needing two unpaid steps cost about 10^20 parasites. Counted per site, all mammals searched as thoroughly as the parasite, so the head-count argument fails and only its per-lineage form survives.

HIV Evolution Published In roughly 100 years, HIV produced about 10^22 mutations across 6 × 10^22 particles but fixed no more than about 5,000 constructive changes. The article argues that this measured output is too small to support the creative power required for mammalian evolution. Influenza A century of sequenced influenza records 10^22 to 10^24 mutations. The search bought about a hundred adaptive substitutions, nine new glycosylation sites, four defeated drugs and three pandemics built from existing segments. No new gene, and a measured ceiling on the glycan gains. Lenski's Long-Term Evolution Experiment Thirty-eight years, twelve flasks and seventy trillion cells produced one new metabolic ability, in one population. Presumptive knockouts supply over half the early adaptive mutations and at least a quarter of the late ones, yet fitness rose 70% through 60,000 generations. The Discovery Institute's stepwise-citrate claim holds up, and every measurement behind it was made by Lenski's own collaborators. Measuring Biological Information Szostak and Hazen's functional information is the sound measure of biological information and cannot yet be computed for a genome. Durston's Fits, Gitt's definition and algorithmic specified complexity give no usable number for one either. This series scores mutations as gain, loss, modification or transfer and counts new nucleotides. Behe's First Rule is right about rates, unproven as a limit. Mycobacterium tuberculosis The tuberculosis bacillus divides about once a day, mutates slowly and exchanges no genes, so every adaptive change arose by mutation inside the lineage. About 10^19 bacilli in the drug era produced resistance to every drug used against them, all by altering or breaking parts they already had. Ancient genomes date the complex to under 6,000 years. Plasmodium falciparum In about 75 years of drug treatment, between 10^21 and 10^22 malaria parasites tried every single-letter mutation billions of times. The documented result is one new transport activity, one deleted gene and a set of modified enzymes, most with a fitness cost. Chloroquine resistance cost about 10^20 parasites because its first steps do not help alone. SARS-CoV-2 The best-sequenced epidemic in history searched every one-letter change to its genome about 10^16 times over and every two-letter pair about 10^9 times. It bought higher receptor affinity, antibody escape, a shifted entry route, one costly drug-resistance path and several knocked-out genes. No new gene, protein or enzyme appeared. When Evolution Creates Information Mutation and selection have been watched building new functions: citrate use, a second phage receptor, new influenza glycosylation sites, a laboratory ligase. The claim that mutations never add information is false. The watched gains are small, at most four new nucleotides or tens of bits each, far below the 500-bit design threshold, and outnumbered by losses and modifications.

How Hard Is a New Function? (3)

How much searching a specific new capability costs, measured where the search was watched from beginning to end.

Antibiotic Resistance Most clinical antibiotic resistance is a gene moved between bacteria or a part of the cell broken, and resistance genes existed before 1940. But the TEM extended-spectrum beta-lactamases are a real gain built by five point mutations, and the fitness cost the creationist argument leans on is real but routinely compensated. Directed Evolution and Protein Engineering Laboratory searches of four thousand to six trillion variants reached new activities within 3 to 23 substitutions, nearly always from a protein that already had a trace of the target. One search made a new catalytic fold, small and zinc-bound. Hall's ebg experiment replaced one deleted enzyme by improving a feeble relative; it did not rebuild a system. Phage Lambda and the OmpF Switch Phage lambda gained a second way into its host by four substitutions in one gene, in twelve days, in 24 of 96 flasks. The four are jointly required but were not jointly acquired: each step improved binding to the old receptor, so selection paid for them one at a time. The change scores as a gain of function.

Observed, or Only Inferred? (2)

Evolution has been watched in small steps within species: a few mutations, or shifts in variants already present. The large gains, such as new genes, fish antifreeze and species radiations, are reconstructed from comparisons. Some famous cases sold as observed were never watched, and several design and young-earth claims, including Axe's protein rarity and genetic entropy, are inferences too.

Nylonase Bacteria that eat nylon by-products were isolated within decades of the polymer, but nobody watched the enzyme NylB arise. Ohno's 1984 frameshift was a proposal that readers reported as a finding. NylB is an older esterase whose nylon activity two substitutions raise: a modification of function, not a new gene. All three nylon enzymes belong to old folds.

Plants and Animals (9)

Most change watched in wild and domestic plants and animals ran on variation already present, as creationists predicted and mainstream reviews agree. The peppered moth ran on a new mutation. None of the watched changes is a gain of function, and the old variation carries ages young-earth models have not answered.

Cichlid Radiations Lake Victoria's 500 cichlid species are timed by a lake that dried out about 15,000 years ago, but new cores show ponds survived in the basin. The species were sorted from old variation mixed by hybridisation, not built by new mutations. Nobody watched one form, and a post-Flood timeline needs rates about three times faster. Darwin's Finches Forty years on Daphne Major measured strong, heritable selection on finch beaks. It reversed three times, and the birds ended smaller-bodied and pointier-beaked, not back where they began. Every change traced to gene variants older than the species or to interbreeding. No new gene was reported. It is the clearest record of selection and weak evidence of new information. Dog Breeds Most of the 350-odd dog breeds were formed in the last 150 to 200 years by sorting variants dogs already carried, some from wolves and many from earlier dog mutations. The breed-defining changes are losses, weakened genes and extra gene copies; none is yet shown to be a new function. The price was inbreeding and disease. Italian Wall Lizards Ten Italian wall lizards moved to Pod Mrčaru in 1971 turned to plants and grew bigger heads and gut valves within about 35 years. No inherited change has been shown: the valves vanished on an insect diet and head differences faded in a common garden, so the case is not yet scoreable as evolution. Observed Speciation New plant species have been watched forming by chromosome doubling, isolated from their parents in one generation. The doubling adds no new gene, and the young genomes then lose genes. In animals only partial isolation has been watched, and the London Underground mosquito did not evolve in the Underground. Peppered Moths Bird predation made the peppered moth mostly black within decades and made it pale again when the air cleaned up. Majerus's 4,864-moth experiment confirmed the predation, and Kettlewell was not a fraud. The black form came from one new transposon insertion, most likely around 1819: a copy that raises an existing gene region's output, not a new function. Sticklebacks Lake sticklebacks lose armour within one to two decades by reusing an Eda allele already present in the sea, and lose the pelvis through different, overlapping deletions of one fragile Pitx1 enhancer. No new adaptive mutation was watched arising. A measured mutation rate dates the reused allele to about two million generations, which young-earth models have not answered. The Creation Model of Rapid Speciation Young-earth models say Ark kinds became today's species by sorting built-in variation. Fast change from stored variation is well supported, and the average kind needs rates about twice the fastest radiation known. The variation's measured age, the four-allele ceiling and the failed tracking prediction remain unanswered.

Selection in Human Populations (3)

Alleles that measurably spread through human populations within recorded or datable time, with their selection coefficients and what each one changed.

EPAS1 and the Tibetan Highlands The Tibetan EPAS1 haplotype came from Denisovans by interbreeding, so it is a transfer, not a new mutation, and it tells nothing about how long a mutation search takes. Its measured effect is a blunted response to low oxygen. Yi's 2,750-year date, once called the fastest human sweep, has since been moved to tens of thousands of years. Lactase Persistence Lactase persistence comes from at least five separate one-letter changes in one enhancer that keep the lactase gene on in adults. Ancient DNA shows the European allele rose mainly in the last 3,200 years. By Behe's 2010 rules the main variants modify an existing element, and such changes are cheap in populations of millions. Malaria Resistance Alleles Five human defences against malaria, measured in living cohorts. Sickle and haemoglobin C change a working protein at one codon; haemoglobin E adds a splice site; G6PD deficiency and Duffy negativity break things, and neither fights falciparum. A glycophorin rearrangement makes a hybrid gene and protects as well as sickle trait.

The Fossil Record (28)

The fossil order is real, and no young-earth model reproduces it. The named transitional sequences are real, and four cross the created-kind line unexplained. Young-earth claims hold for fast burial of single fossils, gut contents, dinosaur soft tissue and unexplained stasis. Nests, every claimed out-of-place fossil and the claim that all phyla appeared at once fail.

The Cambrian Explosion The Cambrian explosion is real. Burrows, which need no special preservation, jump in variety and depth at the base of the Cambrian, and phylum-rank first appearances pile up in its first three stages, then stop. But Ediacaran animals and Cambrian stem groups refute claims that all phyla appeared at once without intermediates.

Claimed Fossil Anomalies (4)

No claimed human track, tool, pollen grain or living dinosaur examined on this site survives four tests: provenance, stratigraphic context, independent identification and open publication. Creationists killed many of these claims themselves, and their organisations have retired them unevenly. The one real anomaly, soft tissue in dinosaur bone, is a preservation puzzle.

Dinosaurs in Ancient Art and Text No ancient or medieval image or text examined here shows a living dinosaur or pterosaur. The Ica stones and Acámbaro figurines are modern, the rock-art dinosaurs are composites and stains, and Ta Prohm, the Carlisle brass and the Nile mosaic show real or mythical animals. Behemoth fits a hippopotamus in hyperbole, or a mythic beast, better than a sauropod. Human and Dinosaur Tracks Together No human footprint has been shown in the Cretaceous Glen Rose Formation. The Paluxy 'man tracks' were carvings sold in the 1930s, erosion marks, and elongated dinosaur tracks whose three toes reappeared as coloured infilling; ICR withdrew the claim in 1986 and its own chemistry found no tampering. The Delk slab and Baugh's museum prints fail the tests ICR itself set in 1976. Living Dinosaur Claims No credible evidence shows a non-avian dinosaur, plesiosaur or pterosaur alive today. Forty-five years of Congo expeditions, Papua New Guinea searches and Loch Ness surveys returned testimony, lights and blurred images, never a body, bone or reptile DNA. The Zuiyo-maru carcass was a basking shark. Creation Ministries International retired the claim in 2018; Answers in Genesis has not.

Reading the Record (7)

The fossil record reads reliably for order and comparisons, poorly for ancestry from shape and restored soft parts. Completeness depends on the group: a missing shelly sea animal is evidence, a missing insect is not. No young-earth model reproduces the order. Rapid burial fits spectacular fossils, not the rocks; frauds discredit nothing; dinosaur soft tissue remains unsolved for deep time.

Dinosaur Soft Tissue Dinosaur bones preserve flexible vessels and protein fragments that several laboratories identify as original, far beyond measured collagen decay times. Iron cross-linking could close the gap only if a two-year, 200-fold effect held for 66 million years in cool ground; no experiment shows that. Radiocarbon in the bones is real but cannot be read as an age. Fossil Preservation Rapid burial is real and mainstream for the fossils creationists cite: articulated skeletons, soft-bodied Lagerstätten, phosphatised tissue. It is not universal. Most fossils are shells, which need no rapid burial and on modern sea floors lie for centuries, and each Lagerstätte records many small burials in one lake, lagoon or shelf. Preservation bias cuts both ways. Fossil Reconstruction and Fraud Fossil frauds were exposed by scientists and do not discredit paleontology. Reconstructions add uncertain soft tissue and behavior, so measured bones must be separated from artistic inference. How Complete Is the Fossil Record? Completeness depends on the group. Shelly marine genera are re-found in about four of five stages they span; insects and spiders in about one in three. Absences among durable sea animals are evidence; absences among insects and soft-bodied animals prove little, and land vertebrates fall between. Denton's 79% of living land-vertebrate families falls to 52% before the Pleistocene. The Order of the Fossil Record Two million records show that fossils occur in a strong global order. Hydrodynamic sorting, escape ability, ecological zonation, and published creationist models do not reproduce it, so the order remains a major problem for Flood geology. Trends in the Fossil Record Stasis is real but weaker and more mixed than usually claimed, and it does not disprove evolution, though nobody has explained its bound. Gaps grow with rank only qualitatively. Convergence is the strongest of the four: morphology proves an unreliable guide to relatedness wherever molecules can check it. Completeness cannot carry the argument usually asked of it.

Stasis, Extinction, and Evolutionary Tempo (3)

Stasis is real but does not decide between the models: deep time must explain why change stops, a young earth why it ran so fast. Living fossils hold at family rank and mostly fail at genus and species. Mass extinctions were real deaths, not burial order, and no young-earth model yet fits the recoveries into the Flood year.

Living Fossils Most famous living fossils are unchanged only at family rank or above. Their living genera are young or fossil-free, and only ginkgo leaves carry a living species into Cretaceous rock. Long stasis is cheap on a young earth and needs active constraint over deep time. The gaps are real rarity, not missed sampling, on both models. Mass Extinctions The Big Five are real in single rock sections: sudden losses within centimetres, then survivors, dwarfs, missing groups and a stepwise rebuilding above. Young-earth models put four inside Noah's Flood. Reading them as mere burial order fails a database test; fitting the recoveries into one year is unsolved.

Trace Fossils (5)

Tracks, nests, dung and bored bones record animals alive on a surface. Oard's BEDS model fits single fast surfaces, and traces whose clock ran inside a living animal cost the Flood nothing. Nests fail: one clutch needed 99 counted days against an 80-day window. Stacked track horizons and bored bones are unfinished.

Briefly Exposed Diluvial Sediments Oard's BEDS model fits a single fast surface, such as a trackway or freshly buried dung. It fails on nests: one Devil's Coulee clutch needed 99 to 171 days on its surface, beyond Oard's 80-day window. Stacks of 160 to 300 track horizons need a walked-on low tide every twelve hours. Nobody has modelled the exposures. Coprolites, Gut Contents, and Feeding Traces A meal in the gut or a healed bite records time inside a living animal, which a Flood model can spend before the Flood, and gut contents support fast burial. Dung worked by beetles, weathered latrines, and insect- or worm-bored bones need a day to two months on each surface. Young-earth writers have priced only the latrines, in words. Fossil Nests and Eggs Dinosaur nesting grounds record months on dry ground: eggs laid in dug nests on soils, hatched and trampled, with three to nine nesting horizons stacked at one site. Floods did bury most clutches, as the young-earth side says, but the 80-day BEDS window cannot hold even the 99 counted days of one Devil's Coulee incubation. Fossil Tracks and Trackways A fossil trackway needs a surface that was firm, shallow or dry, and exposed long enough to set, then buried before rain or trampling erased it. Megatracksites stack such surfaces by the dozen. Undertracks and substrate strength, not swimming, explain manus-only sauropod trails. Tracks before bodies are routine on both timescales, and a footprint rarely names its maker, Laetoli included.

Transitional Sequences (7)

The named fossil sequences are real: each runs in stratigraphic order, and the flat young-earth objections to them fail. Jaw-to-ear, fish-to-tetrapod, whales and birds cross the family line creationists draw for created kinds; the horse series stays inside it. The museum ladders are wrong, real gaps remain, and no young-earth model predicts the order.

Archaeopteryx and the Origin of Birds Archaeopteryx combines bird and dinosaur traits. Creationist and evolutionist statistics both put it on the border, with the raptors in some analyses and the birds in others, so “just a bird” claims too much. Creationist analyses find gaps between most birds and the raptors; its tree position is one parsimony step from the alternative. The Horse Series The horse ladder is wrong; the fossil record is a branching bush with stasis and reversals. Its stratigraphic order is real, but every member lies within Equidae, so the sequence fits creationist ancestry within a created kind, though only with very rapid change after the Flood or a Flood burial order nobody has yet explained. The Missing Transitions Turtle and Romer's Gap transitions have closed. No fossil yet shows how bat, pterosaur or insect wings arose, and specialists concede it. Most fossil genera have gaps in their ranges, so one missing lineage proves little, but a long blank before a whole group needs a specific explanation. The Reptile Jaw and the Mammal Ear Embryology predicted that two reptile jaw bones became mammalian ear bones before the transitional fossils were found. The fossil sequence preserves intermediate double-jawed animals, and the young-earth objections fail. Tiktaalik and the Fish-to-Tetrapod Transition Tiktaalik was found by a real prediction and has transitional anatomy, but older tetrapod tracks falsified its place as an immediate ancestor. The sequence remains real, branching, and full of gaps rather than a straight ladder. Whales and the Land-to-Sea Transition The Eocene fossils preserve a real sequence from four-legged mammals to whales, and the usual young-earth objections fail. The sequence crosses more than one created-kind boundary and remains an unsolved problem for young-earth creationism.

The Origin of Life (13)

Overview of the Origin of Life series. No laboratory has run a replication cycle on the whole of a molecule, or reached a cell, from plausible prebiotic chemistry. Every gap that has closed since 2008 was a monomer-synthesis gap; none in polymerisation, information or replication has closed. Popular probability arguments attack processes no researcher proposes.

Homochirality Life uses left-handed amino acids and right-handed sugars. The claim that nature cannot produce that bias is dead: demonstrated mechanisms amplify a small excess to near purity, most at a steep material cost. Still unsolved is where the first bias came from and why it chose one hand. Fresh asteroid samples are racemic, badly weakening the leading answer. Is the Genetic Code Optimized? The famous result that the genetic code is one in a million is true and misreported by both sides. In the space actually searched, about two trillion codes beat it, and it sits on the slope of a mediocre peak reachable in a dozen swaps. The first laboratory test of the theory behind it failed to find the predicted effect. Making Nucleotides The claim that nobody has ever made a nucleotide died in 2009. Ribose is still unstable, the formose reaction still gives no clean selectivity, and no published route makes all four RNA letters from one-carbon feedstocks in one place. What survives is the chemist's hand: every route needs a fixed sequence of environments nobody has found in the field. Metabolism First and Hydrothermal Vents The main alternative to RNA-first says chemistry came before genes. Iron drives nine of the eleven Krebs intermediates without enzymes, and minerals fix carbon dioxide at millimolar yields. But no self-sustaining cycle has been demonstrated, Orgel's specificity objection is unanswered, and the field's three camps each hold an objection the other two cannot meet. Miller-Urey and the Early Atmosphere The 'wrong atmosphere' objection to Miller-Urey is half right. Hadean zircons say the bulk atmosphere was not reducing, but it was not oxygen-rich either, and since 2008 spark experiments have worked in neutral gas. What survives is stronger: a racemic tar in which monomers are a minority, no polymer, and a neutral sky that delivers little cyanide to the ocean. Panspermia Space delivers amino acids, all five nucleobases, sugars and 33 amino acids in the Bennu sample. It has never been shown to deliver a nucleoside, a peptide or a membrane. Transport between planets is real, but relocating the origin supplies no mechanism, and every version of panspermia needs deep time, so it is no friend of the young-earth position. Protocells and Membranes Fatty acids form vesicles by themselves, feed, grow, divide and compete without any genome. The unsolved problem is stated by Szostak's own laboratory: RNA copying needs 50 to 200 millimolar magnesium and a fatty-acid membrane falls apart above about 4. No protocell yet both divides and copies its contents across generations. The Earliest Evidence for Life Every claim of life older than about 3.5 billion years has a published rebuttal, usually written by geologists. And the Late Heavy Bombardment that made the time window look impossibly narrow probably never happened, which was settled by lunar farside samples returned in 2024. The no-time argument is in worse shape than its users realise. The Last Universal Common Ancestor Reconstructions of LUCA disagree by an order of magnitude, from 355 gene families to 2,657 proteins, and a survey of eight studies found none resembling any other. What all of them agree on is that LUCA already had a genetic code, ribosomes, a membrane and ATP synthase, which makes it a poor stand-in for a first cell. The Minimal Cell The smallest organism ever built has 473 genes and a third of them have no known function. It is not a model of a first cell: it was stripped down from a modern bacterium, booted inside a living cell, and grown in medium supplying everything. The smallest natural genomes are worse models still, because they are parasites. The Ribosome and the Origin of Translation The ribosome's core has a derivable history, and the inferred proto-ribosome has been built and makes peptide bonds. What no laboratory has produced is a process that assigns amino acids to codons. Michael Lynch calls the origin of translation the most enigmatic unsolved problem in the origin of life, and Eugene Koonin reaches for the phrase irreducible complexity to describe it. The RNA World In February 2026 a 45-nucleotide ribozyme made its own complementary strand and a copy of itself, so the claim that no molecule has ever copied itself is dead. What survives is narrower: no whole ribozyme has completed a replication cycle, yields are two parts in a thousand over 72 days, and the feedstock is made by a protein enzyme.

History (123/133)

Ancient Israel: Slavery, Servanthood, and Social Welfare Published Israelite servitude functioned mainly as debt relief and protection from poverty, not race-based chattel slavery. Kidnapping carried a death penalty, escaped servants could not be returned, injury required release, and foreign residents received legal protection, though the system still permitted ownership and unequal status.

Evidentiary Crafts (3/9)

The specialist methods needed to judge manuscripts, artifacts, inscriptions, languages, chronology, and disputed ancient claims.

Archaeology, Stratigraphy, and Destruction Layers Planned How excavators date sites, distinguish occupation phases, identify destruction, and avoid circular use of biblical chronology. Coins, Papyri, and Administrative Records Planned What ordinary documents reveal about rulers, titles, taxes, travel, dates, and the daily setting of biblical narratives. Early Christian Art and Material Culture Planned Catacombs, Dura-Europos, Megiddo, early symbols, crucifixion images, inscriptions, ossuaries, and misleading artifacts. Forgery, Provenance, and the Antiquities Market Laboratory tests show that a stone, sherd or scrap is old far more easily than that its writing is. The Museum of the Bible fragments, the Gospel of Jesus' Wife and the Moussaieff ostraca are very probably modern writing on old material; the Jehoash tablet is probably forged but contested; the James ossuary and pomegranate inscriptions remain unresolved. Inscriptions, Languages, and Writing Systems Planned Paleography, dialect, alphabet development, epigraphy, and why scripts rarely yield an exact year by themselves. Names, Identity, and Ancient Populations Planned Name frequencies, patronymics, ethnic labels, population estimates, and the controls needed before a pattern becomes evidence. Statistics for Ancient Evidence Planned Multiple comparisons, selection bias, missing data, name tests, Bayesian priors, and false precision in sparse historical records. Who Could Read and Write? Reading, signing, writing and composing were separate skills, often split by language. Population literacy rates for Roman Judaea are models. In one selected document sample, Wise classifies many male parties and all ten female parties as unable to sign. Other people's writing reached nonwriters. No rate settles what a named disciple could do.

Ancient Chronology and Calendar Conversion (1)

Regnal years, eclipses, king lists, calendars, synchronisms, and how dates acquire honest uncertainty ranges.

How the Dates of the Hebrew Kings Are Known: Mesopotamian Absolute Chronology Assyrian year-lists anchored by the eclipse of 15 June 763 BC, inscriptions naming nine Hebrew kings, and Babylonian tablets locked by astronomy fix the Hebrew kings' dates to the year from 853 to 539 BC, and the Bible's regnal numbers meet them wherever testable, with two named seams. The soft joints are 587/586, Artaxerxes' Jewish year 1, and inferred coregencies.

Genesis and Early Biblical History (27)

The text, chronology, and physical claims in Genesis examined against manuscripts and external evidence. This page is the front door to the section and links to all of it; the summary and verdict are still to be written.

Ancient Cosmology and Genesis Genesis 1 uses the sky-language of its neighbours. The raqia question is two-sided: the author probably pictured a solid sky, the text never asserts one, and Job 37:18 settles nothing. The flat-earth charge fails on every verse and on the history of the myth. Genesis 2:19 and the creation order trouble concordism, not the plain reading. Noah's Ark: The Logistics The Ark can hold juvenile animals if created kinds are broader than genera, but adult animals do not fit. Ventilation, wooden seam strength, labor, disease, and post-Flood population recovery remain unsolved. The Age of Earth in Church History Nearly every Jewish and Christian writer before 1750 put creation under 6,000 years before his own day. The claim that the church fathers taught long creation days fails on their texts. Deep time entered Christian thought from geology after 1780, not from exegesis, and its first Christian adopters said so. The Genesis Genealogies and Gaps Gaps in the names of Genesis 5 and 11 are possible, and William Henry Green proved that much in 1890. But each verse dates the birth of the son it names, so a missing name adds no years. No surviving text-form puts creation before about 5,600 BC. Gaps cannot reconcile Genesis with an older humanity. The Genre and Grammar of Genesis 1 Judged by Hebrew grammar and genre alone, Genesis 1 presents six ordinary consecutive days: yom with a number, with evening and morning, in a waw-consecutive narrative chain, read that way by Exodus 20:11. The text establishes what its author meant, not whether he was right. The Long Lifespans of Genesis Genesis reports lifespans near 900 years and a sharp decline after the Flood. The numbers are internally consistent and their post-Flood shape is not generic, but Genesis 5 carries an arithmetic signature no real cohort would have, and every proposed biological mechanism fails by one to three orders of magnitude. Which Bible Text Gives the Right Date for Creation? Genesis manuscripts give creation dates about 1,400 years apart. The Masoretic numbers have the best case at about 70 percent confidence; the Dead Sea Scrolls preserve none of the disputed ages, and all three chronologies are young-earth.

Adam, the Fall, and Death (2)

Both Testaments treat Adam as one man, and Paul's argument in Romans 5 and 1 Corinthians 15 does not work without him. Archetypal and mytho-historical readings can be stated fairly; each costs something Paul says. Cain's wife was his sister and the genetic objection fails on arithmetic. The Fall cursed the ground, not physics.

Death Before the Fall Genesis 1:29-30, 3, 9:3 and Isaiah 11 most naturally tie animal death to the Fall but never state it, and Romans 5:12 speaks of human death. Augustine and Aquinas taught pre-Fall carnivory; Theophilus, Irenaeus, Basil, Luther, Calvin and Wesley did not. The reading commits a young earth to placing every fossil predator, tumour and venom after Adam.

Competing Readings of Genesis (3)

Nine readings of Genesis 1 are tested against the same text. The calendar-day reading passes the most textual tests but has weak spots. Every old-earth reading was proposed after the geology it reconciles, by its authors' own admission, and each fails a test the plain reading passes. Passing the tests shows what the author meant, not that he was right.

Gerald Schroeder's Six Days Gerald Schroeder argues the six days of Genesis 1 and the 13.8 billion years of cosmology are one interval seen from two frames, converted by a factor near a trillion. The relativity is real but the conversion is not: it needs a privileged cosmic clock that general relativity does not supply, a fixed ratio that expansion does not give, and a halving series with no derivation. The order of events also fails. Theistic Evolution: A Critique A Christian can accept common descent and keep every article of the ancient creeds. No version keeps Paul's Adam whole: each gives up a first man, a recent sole ancestor, or human death through his sin. Guidance hidden from science is coherent but predicts nothing, so science neither supports nor refutes it.

Genesis and Ancient Near Eastern Parallels (2)

The Babylonian flood and creation tablets share real, specific features with Genesis 1-11, and they are older as written objects. Read against the primary texts, the flood parallel is close, the creation parallel is thin, and literary dependence in either direction has not been shown. The article states what evidence would show it.

Flood Legends Around the World Hundreds of cultures tell of a great flood, and the stories cluster in ways chance does not explain. But a recount of 262 legends kills the percentages apologists quote, the world's commonest flood story is not Genesis-shaped, and every legend that matches Genesis in full is either Mesopotamian or was recorded after missionary contact.

Human History Across the Flood Date (5)

Egypt's early dynasties, the Near Eastern Neolithic, Japan's Jomon and Australia's first settlers lie before or run through every biblical Flood date, on clocks that agree with each other. The Septuagint's date clears Egypt's pyramids but nothing older. No published compression works, and population growth shows a young human race is possible, not that it is young.

Australia, Japan, and the Americas Before the Flood Date Madjedbebe (65,000 or 50,000 years), Lake Mungo (40,000), Japan's first pottery (about 16,000), Monte Verde (14,500) and White Sands (21,000 to 23,000) rest on two or more dating methods that agree. Young-earth writers put them all after Babel, which needs clocks running 3 to 100 times fast. The one published young-earth radiocarbon curve puts White Sands at Babel itself. Egypt and the Flood Date Egypt's Old Kingdom is not dated by the Sothic cycle, so the standard creationist attack moves it zero years. Two independent clocks, king lists and radiocarbon, agree to within 178 years across 1,500 years, and both put the Step Pyramid before every Masoretic Flood date. The Septuagint's Flood date clears the pyramids without touching Egyptology; nothing clears the Predynastic farmers. Human Population Growth Eight people can become eight billion since a biblical Flood at under half a percent a year, so the arithmetic works. But real populations grow in booms and crashes, so an average rate dates nothing. The Stone Age burial count fails the same way: neither side knows how many bodies survive. The Neolithic Revolution and the Flood Date Göbekli Tepe, Jericho and Çatalhöyük sit 6,000 to 10,000 years before any Flood date Genesis allows, in the region Genesis puts Noah and Babel. The Septuagint closes about 11% of the gap. Tell depth alone proves nothing, and the young-earth side is right about that. Measured house lives, counted tree rings and stacked periods do, with no radiocarbon.

Searching for Noah's Ark (5)

Genesis 8:4 names a region, the mountains of Urartu, not a peak, so the Bible requires no particular site and no surviving Ark. No candidate has produced identified wood, sampled independently and published. Every wood date published in full is under 1,700 radiocarbon years; every result said to favour an Ark lacked a named laboratory or a full report.

Mount Ararat and the Ararat Anomaly No evidence puts Noah's Ark on Mount Ararat. The sightings trace to legend and admitted fiction, Navarra's wood radiocarbon-dates to about AD 700, the Ararat Anomaly reads as rock and ice, and the 2010 Hong Kong wood dated young. Young-earth geologists themselves build the volcano after the Flood. Mount Cudi and the Older Tradition From Berossus about 280 BC to the tenth-century Muslim geographers, the oldest landing traditions point to the Kordyaean mountains on the upper Tigris, where Cudi Dağı stands, not to Ağrı Dağı. Most of them localise Genesis 8:4, Berossus attests only a relic cult, and nothing found on Cudi is from a ship. Other Proposed Landing Sites No landing site proposed besides Ararat, Durupınar and Cudi holds up. Creationist geologists read Cornuke's Iranian 'beams' as rock, and his own dated wood is late medieval. The other candidates rest on readings, not finds. Boat-shaped flow forms recur around Ararat, so Durupınar's shape is not evidence. Of eleven candidates, none tested has survived. The Durupınar Formation The boat-shaped Durupınar formation is a natural landslide feature, not Noah's Ark. It is about twice as wide as the Ark, the 1988 cores found mud and rock but no wood, and every sectioned 'wood' sample was volcanic rock. Creationist scientists debunked it first. The 2019-2026 radar and soil claims are unpublished.

The Flood Narrative (2)

Genesis 6-9 describes a flood that covers every high mountain, kills every land creature outside one boat, and is sealed by a promise never to repeat it. Its author meant a worldwide flood, and its most capable old-earth readers concede this. The local-flood reading was invented in the 1650s. The source division is a real pattern but a weak argument. Whether the flood happened is a geology question this article does not settle.

The Sons of God and the Nephilim Genesis 6:1-4 has three readings. The angelic reading is the oldest, the only one attested for four centuries until Jewish translators offered a human reading in the mid-second century AD, and it has the best lexical evidence, since the exact phrase 'sons of God' names heavenly beings in all five of its other uses. The Sethite reading is a third-century Christian rescue with textual problems. The royal reading is a second-century rabbinic idea revived by Kline. The 120 years of 6:3 is a countdown, not a lifespan. Nothing on this site depends on which reading is right.

Babel, Language, and Human Dispersal (14)

Genesis 10 and 11 describe real Mesopotamian building and real ancient peoples, and the universal reading is the natural one. But Shinar's temple levels and two written languages, the ages of the text-dated language families, and Africa's genetic depth exceed what a Masoretic Babel allows. The Septuagint brings the word-change test within reach and leaves the genetic gap.

Language Families and the Tower of Babel (4)

Glottolog counts 421 families and isolates with no demonstrated links, as Babel predicts but also as the comparative method's horizon predicts on any chronology. Text-dated Indo-European and Sino-Tibetan are 6,000 to 8,000 years old. Under a Masoretic Babel, Hittite and Vedic need 2.5 to 4.8 times the fastest recorded rate; the Septuagint date with small founder groups nearly fits.

Did Language Begin Once? Linguistics cannot show whether language began once or several times. No link between the world's 421 families is demonstrated, but the comparative method loses its signal after 6,000 to 10,000 years, so Babel and a long history predict the same. Proto-World, the macro-families and the phoneme gradient fail statistical tests, and Proto-World contradicts the young-earth model. How Old Are the Language Families? Indo-European and Sino-Tibetan, dated on written texts alone, come out 6,000 to 8,000 years old, with lower bounds older than every Masoretic Babel except one Sino-Tibetan bound. Deep families without old texts rest on radiocarbon. On conventional dates Egyptian and Sumerian are written before every Babel date in print; young-earth revisions remove this only by lowering all ancient history. Is There Time Since Babel for the Language Families? Documented lineages keep 70 to 85% of their basic vocabulary per millennium. Calibrated on 59 text-dated Indo-European pairs, Hittite and Vedic need 2.45 times the fastest recorded rate under a Masoretic Babel, 3.5 to 4.8 with rate variation. Small founder populations halve the gap; the Septuagint date then nearly fits, and Austronesian cannot exclude a post-Babel origin.

The Table of Nations (5)

Genesis 10 name by name: 32 names are secure, 7 probable, 15 contested and about 30 unattested. Its main Japhethite names carry 9th-7th-century Assyrian forms, yet it omits Persians, Chaldeans, Arabs and Urartu. The three lines are not language families or genetic clusters. The best fit is an old core updated in the Iron Age, not a one-off post-Babel record.

Genesis 10 Name by Name Of the names in Genesis 10, 32 are secure, 7 probable and 15 contested, and about 30 are unattested outside the Bible. The secure core is in the right places and combinations. Nimrod is unattested, and the popular matches with Moscow, Tobolsk, Germany and the Denisovans rest on sound alone. How the Table of Nations Has Been Read From Jubilees to the medieval pedigrees, readers matched Genesis 10 to their own maps, so reception adds no independent evidence. The English royal pedigrees reach Noah only in about 892. The young-earth matches with modern nations rest on sound alone, and Carter's 2024 overlapping lines are the best young-earth reading but leave little to test. Testing the Table of Nations Japheth, Ham and Shem are not language families, and they carry no genetic signal, though the chapter's ethnography holds where ancient DNA can test it. Two generation counts strain the Masoretic dates and fit the Septuagint better. The Table's map is drawn from the Levant, not from Shinar. When Was the Table of Nations Written? Genesis 10's names were first recorded over two thousand years. Its core Japhethite names first appear in Assyrian texts of 837-669 BC, in Assyrian spellings, yet it omits the Persians, Chaldeans, Arabs and Urartu. That fits an old core updated in the Iron Age, as Kitchen and Ross hold, and not a one-off post-Babel record.

The Tower of Babel (4)

Genesis 11:1-9 is authentically Mesopotamian in its details and was read by every ancient reader as all mankind. But no Mesopotamian text or layer records a tower-and-tongues event; the temple platforms of Eridu and Uruk precede every genealogy-based date; AiG's 2242 BC has no sound source; and the Masoretic text allows a Babel up to 340 years after the Flood.

The Tower of Babel and the Mesopotamian Texts No Mesopotamian text tells the story of Genesis 11. Its building details are right, down to the idiom, but the Sumerian spell has no tower, K 3657 is about something else, the closest parallels are eighth- and sixth-century royal texts, and the world legends most like Genesis were recorded after missionaries arrived. When Was the Tower of Babel? If Peleg's division is the dispersion, Babel falls 101 to 340 years after the Flood on the Masoretic text, and the oldest readers did not choose the first year. AiG's 2242 BC has no sound source. At every date allowed, Shinar already had temples, cities and writing, and on every Masoretic date two written languages. Where Was the Tower of Babel? Genesis 11:3 describes the real building technique of southern Mesopotamia, and Babylon fits every clue in the text; its early levels lie under the water table. But true ziggurats begin under Ur-Nammu, after every Masoretic Babel date, Eridu's twenty levels must fit in about 130 years on AiG's date, and the three young-earth sites exclude each other.

Israelite Religion in History (13)

Israel's religion did change: the wording of God's uniqueness sharpened, afterlife and Satan became clear late, and the prophetic books grew. But the evidence does not support Yahweh as El's son, Yahweh-alone as an exilic invention, borrowed theology or a forged Deuteronomy. It fits a demanded faith many Israelites kept leaving, though nothing outside the Bible reaches Sinai.

Child Sacrifice and the Darker Cult Judahite kings and commoners sacrificed children by fire, and some believed Yahweh wanted it. The Bible records this to condemn it: its laws forbid it, Jeremiah denies God commanded it, and Isaac is spared. Ezekiel 20:25 remains hard, Jephthah probably killed his daughter, and no archaeology from Israel confirms the practice. Deuteronomy and the Pious-Fraud Charge De Wette never called Deuteronomy a fraud, and nothing shows a scroll forged in 622 BC. Josiah most likely enforced an older law, since centralization predates him. But a later narrator framed the book and parts echo a 672 BC Assyrian treaty, so the evidence cannot show Moses wrote it all. Did Israelite Monotheism Evolve? Many Israelites worshipped other gods, and the wording of YHWH's uniqueness grew sharper by the exile. But YHWH as El's son rests on one disputed couplet, names show Judah invoking YHWH almost alone from 800 BC, and the critics' own scholars date the demand for YHWH alone before the exile. How Were Joshua to Kings Put Together, and How Freely Did Chronicles Rewrite Them? Joshua to Kings reached final form after 562 BC from older written records: royal annals it names, cross-dated reigns, notes written while the temple stood, and kings confirmed by Assyrian and Babylonian texts. Chronicles copies closely where it copies, but omits, adds and changes freely; some differences come from another Samuel text. Israelite Religion on the Ground Archaeology matches the Bible's own picture of Israelite religion: Yahweh paired with Asherah, shrines with standing stones and incense, household figurines, short-lived reforms, and a mixed Yahweh cult among Jews in Egypt. The match in kind and date answers the late-invention charge, but cannot show whether Yahweh-alone worship came first. The Late Development of Afterlife and Satan Resurrection, judgment of the dead, named angels and Satan as chief of evil spirits do become clear late, mostly after 300 BC. But each grows from Israel's own texts in Hebrew words, every evil power stays God's creature, and the Persian parallels are late or differently shaped, so the evidence fits progressive revelation better than borrowing.

Old Testament Miracles as History (2)

Assyrian records confirm Sennacherib's failure to take Jerusalem, and a Tyrian record a drought under Jezebel's father. Jericho fell, but too early for either Exodus date. Natural models of the plagues, sea and manna need perfect timing. The NASA missing day and James Bartley's whale are legends to drop.

Why Did Sennacherib Fail to Take Jerusalem? Assyrian records written a year after 701 BC confirm that Sennacherib wrecked Judah, blockaded Jerusalem, took Hezekiah's tribute and, unusually, left him on his throne; a Babylonian chronicle confirms a son later killed him. Why he left is not recorded. The 185,000 dead rest on the Bible alone.

Prophecy and Apocalyptic as Genres (2)

Prophetic books grew after their prophets. Jeremiah's longer edition adds a promise after it came true, and Isaiah 40-55 probably comes from the exile. Apocalypses wrote past history as prophecy under ancient names. These findings remove some predictions from the evidence, but not texts copied before the events they describe.

Who Wrote Isaiah 40-66? Isaiah 40-55 most probably comes from a prophet among the exiles in the 540s BC, and 56-66 from after the return. The strongest arguments for that date do not assume prediction is impossible, but they are not proof. The naming of Cyrus drops out of the evidence for prediction; Isaiah 53 against the Gospels does not.

Ugaritic and Ancient Near Eastern Parallels (2)

The parallels are real. Proverbs 22:17-23:11 drew on the Egyptian Instruction of Amenemope, and Isaiah 27:1 shares a rare phrase with a Ugaritic myth. Psalm 29, the law codes and Deuteronomy's curses show a shared world, not proven copying. No case shows borrowed theology: the differences are pointed. On dates the evidence pulls both ways.

Did Proverbs Copy an Egyptian Book? Proverbs 22:17-23:11 drew on the Egyptian Instruction of Amenemope: 21 of its 24 verses have parallels, it borrows the Egyptian book's 'thirty', and Amenemope is older. The Hebrew writer selected, rearranged and re-founded the sayings on the LORD. The section calls itself 'words of the wise', so borrowing fits inspiration, and it does not fix the section's date.

Old Testament History and Archaeology (14)

Israel’s history tested against its texts, inscriptions, chronology, archaeology, and neighboring records. This page is the front door to the section and links to all of it; the summary and verdict are still to be written.

Archaeological Confirmations of the Old Testament Fifty-three people named in the Hebrew Bible have been identified in ancient inscriptions, on a published criterion of three matching identifying marks. What that establishes is narrower than it sounds, and the honest inventory includes the claims that failed. Archaeology of the First Temple No securely identified remains of Solomon's temple have been excavated. Limited finds from the Temple Mount and an Arad letter fit a late monarchic Yahweh temple, while regional temples establish the biblical plan's plausibility. They do not identify Solomon as builder, date the text, or authenticate the disputed inscribed objects. Biblical Minimalism and How to Test an Ancient Text Neither side of the minimalist debate applies its own standard consistently. This article proposes six tests drawn from cases where the evidence actually moved — including three famous arguments retracted since 1976, two of them by the scholars who made them. David, Solomon, and the United Monarchy David existed and founded a dynasty; two ninth-century inscriptions name his house. The dispute is over scale. Jerusalem's 103 new radiocarbon dates show tenth-century settlement but date its monumental expansion to the ninth century, and the empire of 1 Kings 4 has no archaeological support. Exile, Babylon, and Persia The exile is the best-attested period in Old Testament history. A Babylonian chronicle dates the capture of Jerusalem to March 597 BC, ration tablets name Jehoiachin in Babylon, and a Judean settlement appears in cuneiform. But the Cyrus Cylinder never mentions the Jews. Jericho, Ai, and the Conquest Narratives Joshua says Israel burned only three cities and occupied the rest intact, so the conquest was never going to leave many destruction layers. Jericho's date remains unsettled and currently favors the sixteenth century; Ai has no clean solution; Hazor burned, but who burned it is disputed. Moses and the Composition of the Pentateuch The classic Documentary Hypothesis has fractured, and its linchpin — a seventh-century Deuteronomy — is now contested in mainstream journals. But the Pentateuch also says Moses died, and no defensible position has him writing all five books. The Emergence of Early Israel Egypt names a people called Israel in Canaan by about 1208 BC, and hundreds of new villages appear in the highlands at the same time. But every proposed marker of Israelite ethnicity has failed under testing, including the pig bones, which Finkelstein's own team retracted. The Historicity of Esther Esther's Persian setting checks out in detail: Xerxes' calendar, the palace gate at Susa, the royal post, court customs and names of the period. No character but Xerxes is attested, Marduka was a common name, and Amestris, Xerxes' wife at Susa that winter, is the one hard conflict. The rest is a judgement about genre. The Patriarchs in Their Ancient Setting The Nuzi parallels that once dated Abraham to the second millennium have collapsed, conceded by conservative scholars. What survives is weaker but real: customs unknown in first-millennium Israel, Amorite-type names, and the absence of horses where a later writer would have put them. When Did the Exodus Happen? Dating the Exodus The Exodus cannot presently be dated. The early date better fits 480 years, Jephthah, and Levite generations; the late date fits Pi-Ramesse, 19th-Dynasty names, and Israel in Canaan by 1208 BC. Each makes an anchor text nonliteral, and neither has a settling artifact. When Was the Book of Daniel Written? Daniel's court tales preserve sixth-century Babylonian knowledge and early Aramaic, while its visions contain second-century detail and fail where events move past 164 BC. The evidence supports a layered book and does not settle one date for the whole. Where Were Sodom and Gomorrah? No excavated site has been shown to be Sodom. The southern candidates burned 300 years too early and not at the same time as each other; the northern candidate burned 300 years too late, was rebuilt afterward, and its airburst paper was retracted in 2025.

Prophecy (14)

Tested from the manuscript date forward, Isaiah 53 and Daniel 9 survive as evidence a skeptic must weigh, and Nahum on Nineveh is the best case among the nations. Most popular proof-texts fail. Ezekiel's Tyre and Egypt failed as worded, Daniel 11 goes wrong at Antiochus IV's death, and every dated end-time forecast by later interpreters failed.

Daniel 2, 7, and 11 Daniel 11 tracks Persian and Hellenistic history closely down to Antiochus IV's persecution of 167–165 BC, then fails on his last campaign and death. Daniel 8 names Media-Persia and Greece; the fourth kingdom of chapters 2 and 7 is more likely Greece than Rome, and neither reading yields a checkable fulfilled prediction. Daniel 9 and the Seventy Weeks Being rewritten The article dates Artaxerxes's order to rebuild Jerusalem to 444 BC, treats Daniel's 69 weeks as 483 years of 360 days, and reaches the spring of AD 33. It argues that this predicts Jesus's death, while assigning the final half-week to the future. Eschatology and Israel Jesus's prediction of the temple's fall came true, but it rests on Mark's disputed date and was within a shrewd forecaster's reach. "This generation" either failed or meant only the temple, and the text does not settle which. Every dated end-time forecast since has failed, and Israel in 1948 fails the out-of-reach test. How to Test a Fulfilled Prophecy A prophecy is evidence only if it was written before the event, committed to an outcome that could fail, lay beyond anyone's power to arrange, and survives a count of the misses. Applied evenly, these tests reject Nostradamus and ancient after-the-event oracles, and also Stoner's odds and several favourite Christian proof-texts. Isaiah 53 Isaiah 53 securely predates Jesus and most naturally describes an individual suffering for others. It does not mention crucifixion or generate Jewish belief in a crucified Messiah, but Gospel writers could shape details around it. It is real but inconclusive evidence, not a fulfillment that “could not have been faked.” Messianic Prophecy: An Overview Of thirty Old Testament texts applied to Jesus by the New Testament or popular apologetics, two survive as evidence a skeptic must weigh, Isaiah 53 and Daniel 9, and Micah 5:2 and Genesis 49:10 add a little. The rest are typology, not predictions in their own setting, or matches within reach of Jesus, his followers or the Gospel writers. Psalm 22: "Pierced" or "Like a Lion"? The Masoretic text reads “like a lion”; ancient witnesses support “dug” or “gouged,” and no manuscript says “pierced.” The New Testament alludes to Psalm 22 but never quotes that verse. The psalm predates Jesus, yet Gospel shaping remains possible, so it does not securely predict pierced hands and feet.

Messianic Birth Texts (2)

Read in their own settings, Isaiah 7:14 promised Ahaz a near-term sign and was never read as messianic before Christianity; Micah 5:2 really expected a ruler from Bethlehem, but a birthplace is easy to supply; Genesis 49:10 was read messianically, but the claim that its scepter departed in AD 7 misstates its sources.

Isaiah 7:14 and the Virgin Isaiah 7:14 gave King Ahaz a sign with a deadline, met when Assyria removed both enemy kings by 731 BC. Almah means a young woman, not specifically a virgin, and no Jew read the verse as messianic. Matthew's 'fulfil' is typology, not a claim of prediction, so the verse carries no weight as evidence of foreknowledge.

Prophecies About Nations and Cities (2)

Nahum's oracle on Nineveh is the strongest case: written before 612 BC and fulfilled in its main clauses. Babylon fell to the Medes but was not destroyed. Ezekiel himself records that Nebuchadnezzar failed at Tyre, which still stands, and Egypt was never desolate. Cyrus by name rests on Isaiah's disputed date.

Did Ezekiel's Prophecy Against Tyre Come True? Not as worded. Nebuchadnezzar besieged Tyre for thirteen years without taking it, as Ezekiel 29:18 admits. Alexander's causeway matches one clause, by another agent, and Tyre was rebuilt and is a city today. The book kept its failure on the record, which supports its early date.

Second Temple Messianic Expectations (2)

Dated manuscripts show that Jews before Jesus expected a conquering son of David, often beside a priestly messiah and a prophet like Moses, and sometimes a heavenly judge. No pre-Christian text has a messiah who suffers, dies and atones; every claimed example fails on its reading or its date.

Was Jesus One of Many Would-Be Messiahs? Sources name eighteen Jewish royal, prophetic or rebel leaders between 4 BC and AD 135. None is reported to have retained a following proclaiming him risen. Some movements survived under relatives; other ancient and later groups kept faith in their departed leader. The Christian resurrection proclamation differs from many comparisons, but survival alone establishes no miracle.

Second Temple and Greco-Roman Context (6)

The Jewish and Roman worlds that shaped the New Testament and supply external checks on its people, practices, and chronology.

Ancient Citation and Authorship Conventions Ancient quotation permitted paraphrase and blended wording, but particular source names still need explanation. The texts examined condemn deceptive authorship without establishing how every first audience read pseudonymous works. Authorised composition was possible. Disputed letters require individual evidence, and stylistic tests need validation for their particular text lengths, genres and candidate authors. Intertestamental Judaism The "400 silent years" describe a gap in the Protestant canon and a Jewish tradition about the end of prophecy. Histories, scrolls, papyri and buildings document a shared Jewish way of life alongside disputes over priesthood, calendar, law and resurrection. The famous parties are known mainly through outsiders. Oral Tradition and Memory Paul's letters attest Jesus traditions and repeated formulas by the 50s. Shared Greek wording and varied Gospel narratives support preservation alongside editing, without uniquely identifying each passage's transmission route. Memory studies and the Hogg comparison show how both facts and errors can persist. Historical claims require evidence beyond stability, vividness or agreement alone. Roman Law and Crucifixion Roman prefects held capital authority in Judea; Jewish council powers beyond the Temple remain disputed. Jesus' crucifixion under Pilate is strongly attested, while individual trial scenes are less secure. Crucifixion varied, burial was possible and is known from Judean remains, and citizenship offered protections that officials sometimes violated. The Jewish Revolts and the Parting of the Ways Coins, camps, letters, inscriptions and burned ruins confirm the Roman victories of AD 70, 73 or 74, and 135, though Josephus still supplies most details. No single council, curse or edict separated church and synagogue. The evidence shows uneven local separations from the 50s to the fourth century. Travel, Networks, and the Testability of Acts Inscriptions and ancient texts support many of Acts' provincial titles, routes, ports and sailing details. Gallio's proconsulship, most likely AD 51-52, anchors its Pauline chronology. These checks support a well-informed author without establishing his presence or every narrated event. Theudas, the Jerusalem visits and the independence of some corroborating accounts remain disputed.

Jesus and the Gospels (27/30)

The historical case for Jesus and the tests that can be applied to the Gospel traditions. This page is the front door to the section and links to all of it; the summary and verdict are still to be written.

Ancient Historiography and Compositional Convention Most Gospel differences are what honest independent testimony looks like: courts and memory studies agree the core holds while details drift. Literary dependence explains a documented handful more. 'Compositional device' explains few, and Plutarch never called his own variations devices. Six hard cases remain unresolved. Chronology of Christ Being rewritten Jesus likely began his ministry in the second half of AD 29, continued for a little over three years, and was crucified on Friday, April 3, AD 33. The date combines Tiberius's fifteenth year, John's Passovers, the weekday, and lunar evidence. Gospel Name Frequencies The 79 Gospel men match Palestinian names at 10^36 to 1 against ignorance and reject the diaspora at p = 4.29 × 10^-13. This supports Palestinian tradition, not eyewitness authorship: apocryphal texts fit similarly, and names cannot prove each person historical. Jesus in Archaeology and Outside Sources Josephus's James passage names a Jesus called Christ; Tacitus says Christus was executed under Pilate; both are very probably genuine. Josephus's longer passage probably has a core whose wording cannot be recovered. Later texts mostly report Christians. Archaeology confirms officials and places, not events; no object is securely tied to Jesus. Sacred Sites, Relics, and Pilgrimage Memory Planned The Holy Sepulchre, Garden Tomb, Bethlehem, Capernaum, continuity of veneration, relic fraud, and legend growth. The Synoptic Problem Matthew, Mark and Luke share long runs of identical Greek — measured here at up to 33 words — so at least two of them copied a text. Mark was almost certainly first, which is the strongest evidence against the apostle Matthew writing Matthew. John shows no copying at all. Undesigned Coincidences Biblical accounts contain details that fit without noting the connection. The strongest support shared memory; copying and design explain weaker cases. When Were the Gospels Written? External witnesses set latest dates, while most precise Gospel dates remain disputed inferences. Markan priority links Matthew and Luke to Mark; John has separate dating arguments. The ending of Acts supports an early date only as part of a wider case, and a late Acts does not by itself date Luke equally late. Who Wrote the Gospels? All four Gospels are formally anonymous, and even conservative specialists grant that the titles were added later, by other hands than the evangelists'. The attributions still hold up unevenly: Luke and John have real support, Mark rests on one named early witness, and the apostle Matthew almost certainly did not write Matthew.

Did Jesus Exist? Jesus Mythicism Audited (1/2)

Paul's letters, an early creed, Josephus, Tacitus, and hostile sources make a historical Jesus who died under Pilate far more likely than a celestial one. Mythicism survives only by dismissing each source separately; the Resurrection is a separate question.

Joseph Atwill's Caesar's Messiah Planned Atwill's claim that the Flavian court wrote the Gospels as a satire of Titus's Judean campaign. The proposed parallels with Josephus tested for whether they exceed chance, and the reasons both mainstream scholars and mythicists such as Richard Carrier reject the theory.

Early Christology and Paul (3)

How early Christians came to call Jesus divine and whether Paul changed the message Jesus taught.

Did Paul Write the Disputed Letters? Judged one by one, the six give different answers. 2 Thessalonians probably came from Paul in his lifetime, and Colossians leans that way. Ephesians is unresolved. 1 Timothy and Titus are probably not in Paul's own words, and whether he authorised them is unresolved; 2 Timothy is open. Early Belief in Jesus' Divinity Paul's letters (about AD 48-62) attest prayer to Jesus, application of Yahweh texts to him and probable belief in his role in creation. This rules out a fourth-century invention of his divine status. Inclusion in Israel's one God fits the evidence best, while an agent reading remains viable. The dates do not establish that the belief was true. James versus Paul James and Paul reject inactive belief and give deeds a role at judgment. Their justification formulas clash, but a compatible reading is probable: James credits believers' beginning to God's choice and describes faith completed in action, while Paul excludes deeds as the basis of acceptance. Rahab leaves this less explicit. Paul very probably inherited and developed the core proclamation.

Hard Sayings and Gospel Differences (1)

The return within a generation, divergent resurrection accounts, criteria of authenticity, and when harmonization becomes strained.

The Criterion of Embarrassment A detail that troubled early tellers shows how a story was received and, with more evidence, that it came from tradition; it does not by itself show the event. Weighed so, Jesus' baptism by John is probable, a core of Peter's denial is probable, the women gain little, and the cry's wording is not established.

Is New Testament Scholarship Unusually Skeptical? (2)

Classicists accept Xenophon's authorship of the Anabasis on one witness writing 470 years later and reject the Historia Augusta despite six named authors, so the age of an attribution decides almost nothing. By that standard New Testament scholarship is not unusually skeptical — with one real exception, and four documented cases of conclusions outliving the arguments that produced them.

Bart Ehrman's Main Arguments On the nine arguments tested, Ehrman's factual premises mostly hold: scribes sometimes changed texts for doctrine, ancient forgery was condemned, and some Gospel differences remain unresolved. Several conclusions outrun the evidence: his case against inspiration, forgery across all six disputed Pauline letters, John's changed date, Paul's angel Christ and the 'by definition' miracle argument. Jesus' burial and expectations remain open.

The Nativity Accounts (1/2)

Quirinius, the two genealogies, the Star of Bethlehem, and whether Matthew and Luke can be reconciled.

The Census of Quirinius Luke sets his birth story under Herod and dates its enrollment by Quirinius. The only independently dated Quirinius census is Judea's of AD 6/7, years after Herod died. An earlier Syrian post has disputed indirect support, but no outside record dates an enrollment then. A dating error or conflation is more probable, with moderate confidence. The Star of Bethlehem Planned Matthew 2 and the proposed natural candidates: the planetary conjunctions of 7 BC and 3 to 2 BC, comets, and the novae in Chinese records. What each candidate explains, what it cannot, and whether the text describes a natural object at all.

The Resurrection (10)

The burial, empty tomb, appearances, earliest creed, conversion claims, and competing explanations.

Shroud of Turin: 1988 Carbon Dating Published The 1988 laboratories dated one corner of the Shroud to AD 1260–1390. The article argues that historical warnings, dyes, cotton, gum, vanillin, and chemical differences show that the sampled corner contained a medieval repair, so its date cannot establish the cloth's age. The Resurrection Belief: Event or Process? The resurrection claim, with named witnesses, is early and already dates a bodily raising, but its time profile cannot separate an event from a short burst of visions. Paul's "last of all" closes a list, not all experiences, and closure fits routinization. The thin record of John's followers raises a question for generic grief explanations but cannot decisively test them. The Shroud of Turin Being rewritten The Shroud bears anatomically detailed crucifixion bloodstains and a superficial, three-dimensional body image without detectable paint. The article argues that its cloth, dust, chemistry, image formation, and disputed dating identify it as Jesus's burial cloth and that no known natural or artistic process explains the image.

Natural Explanations of the Resurrection Claims (4)

Hallucination, legend, theft, wrong tomb, swoon, and cognitive dissonance compared against the surviving evidence.

Hallucination Theories of the Resurrection Grief-related experiences of the dead are common and could explain a single early appearance report, but nothing shows that they did. Guilt as a trigger is unproven, grief has no foothold in Paul, and no examined case matches the reported group scenes. Claims that group hallucinations are impossible also fail. The surviving evidence leaves the cause undetermined. Legendary Development Paul records early appearance claims and names men he knew, but gives no scenes. Matthew and Luke visibly rework Mark; whether additions preserve memory or reflect invention needs separate evidence. Physical proof becomes more prominent from Mark to Luke. Later attestation and apologetic framing warrant caution without proving invention. The Swoon Theory A crucified man taken down alive could survive with care, but Jesus very probably died on the cross. No surviving ancient source says otherwise, and every survival version needs unrecorded steps. Medicine cannot diagnose his death. Strauss's objection weighs against a slow recovery; short survival explains the tomb no better than reburial. The Twin and Impostor Theories The evidence gives no clear historical support to a hidden twin or mistaken messenger. Substitution accounts are late theological reports; Qur'an 4:157 names no substitute. Misidentification fits some narratives, but the early appearance list gives little detail for testing it. Comparing these accounts with resurrection requires both prior probabilities and how well each predicts the evidence.

The Historical Case for the Resurrection (3)

The empty tomb, burial, appearances, early creed, women witnesses, Paul, James, and the minimal-facts method.

Did the Apostles Die for Their Claims? Early sources support violent deaths for Peter, Paul, James son of Zebedee and James the Lord's brother, and Paul's letters show years of chosen suffering. No early source records an offer of life for recanting; later legends do. Most other deaths are unknown. This makes knowing fraud by these leaders unlikely, but cannot exclude sincere error. Paul's Letters as Historical Evidence Paul's seven undisputed letters, written about AD 48-62, preserve a resurrection formula older than his preaching at Corinth and report his meetings with Peter and Jesus' brother James. They establish early resurrection belief, a thin account of Jesus' life and Paul's costly commitment, while leaving the appearances' cause and the formula's exact date uncertain. The Empty Tomb Jesus was probably buried, and burial by Joseph of Arimathea in a known tomb is more likely than not. The women's discovery of an empty tomb is a tentative judgment. The guard story, stolen-body report and Jerusalem argument prove less than claimed, and the tomb alone cannot establish a resurrection.

Text and Transmission (4/5)

What the surviving manuscripts establish about the wording and development of the biblical books.

Hebrew Poetry and Literary Architecture Planned Parallelism, meter, acrostics, chiasm, inclusio, wordplay, literary unity, and controls against seeing patterns everywhere. How Textual Criticism Works Editors weigh copies by descent, not by count, and every school accepts that in principle. Seven worked verses show no single rule always wins. The Byzantine and critical texts disagree at three, and two of those remain open. The Textus Receptus departs from every early Greek witness at Revelation 22:19. Counting cannot settle the Byzantine text's history. New Testament Manuscripts & Variants About 5,300 Greek manuscripts preserve the New Testament better than any ancient text. Roughly 40 verses remain doubtful, no doctrine depends on them, and the unobserved first century of copying remains the main gap. The Biblical Canon The four Gospels and most of Paul's letters were used as Scripture by about 200, before surviving council lists. Bishops already judged rival books. Disputes over seven New Testament books and the Old Testament's edges lasted centuries; councils later imposed binding lists, and churches still differ.

Inerrancy and the Missing Autographs (1)

What inerrancy claims, how it handles discrepancies and quotations, and whether a claim about lost originals can be tested.

Bible Contradictions Of 629 alleged Bible contradictions from three skeptical lists and the scholarship, merged into 266 cases, 109 show no real conflict or have a text-supported answer, 113 have a possible but unproved answer, and 41 are hard: 34 unresolved, 4 strained, 3 probable errors.

Sociology (3/37)

Biblical Ethics (2/11)

The moral objections raised against biblical law, narrative, judgment, and Christian teaching.

Abortion, Personhood, and Embryos Planned When human life begins, what biblical texts say, and the moral arguments about abortion, IVF, and embryo loss. Abuse and Corporal Punishment in the Church Planned Institutional cover-up, accountability, and how biblical discipline language has been used to defend harm. Hell and Penal Substitution as Moral Objections Planned Whether eternal punishment and punishment in another's place can be reconciled with justice. The Image of God and Human Rights Planned Whether Christian doctrine grounds equal dignity, how that claim developed, and whether believers behave more morally. Women and Sexuality in Old Testament Law Planned Captive marriage, rape laws, property, polygamy, and whether the legal texts protect women or treat them as possessions.

Purity, Punishment, and Care for the Poor (0/1)

Food laws, capital crimes, gleaning, jubilee, and treatment of resident foreigners in their ancient legal setting.

Harsh Penalties in Old Testament Law Planned The death penalties and physical punishments of the Torah compared clause by clause with the laws of Hammurabi, the Middle Assyrian Laws, and the Hittite laws. Which penalties are milder than their neighbours', which are harsher, and what is known of how often they were carried out.

Sex, Gender, and Marriage (1/3)

The biblical claims and the best medical, psychological, and social evidence bearing on them.

Homosexuality and the Bible Planned The relevant biblical texts, their ancient setting, revisionist readings, natural-law arguments, and the distinction between conduct and orientation. Problems in LGBT Studies Published The article documents biased sampling, inflated statistics, selective publication, activist control of professional bodies, retaliation against dissenters, and poorly supported treatment claims. It concludes that political pressure has degraded parts of LGBT research, while individual studies still require case-by-case judgment. Transgender Identity and Gender Dysphoria Planned Biblical anthropology, biological sex, gender identity, dysphoria, transition outcomes, medical evidence, and competing ethical duties.

Slavery and Servitude (0/1)

Israelite, Greco-Roman, and modern slavery distinguished before the relevant biblical texts are judged.

Christianity, Slavery, and Human Rights Planned Christian arguments on both sides of slavery, abolition, civil rights, legal reform, and the origin of modern rights language.

War and the Canaanite Conquest (1)

The texts, archaeology, ancient war rhetoric, and the moral problem posed by commands of destruction.

Ancient Israel: Morality of the Conquest of Canaan Being rewritten The article argues that Canaanite abuse and attacks warranted judgment, while total-destruction language was often rhetorical and civilians could leave or join Israel. Its defense depends on those textual limits rather than unrestricted violence.

Christianity's Effects on Society (0/6)

Claims that Christianity reshaped Western morality, institutions, learning, art, and everyday assumptions tested against history and rival causes.

Christianity in Art, Music, and Literature Planned The scale of Christian influence on culture, the mechanisms that produced it, and whether cultural reach supports religious truth. Did Christianity Remake Western Morality? Planned Tom Holland's Dominion thesis tested against Greco-Roman morality, Jewish inheritance, Christian teaching, institutional practice, and later secular reform. Jesus as the Most Known Person in History Planned J. Warner Wallace's Person of Interest thesis: an executed Galilean labourer with no army, no office, and no writings became the most widely known human being who has ever lived, which is what the Old Testament said the Messiah would be. The fame is measured, the alternative causes are weighed, and the causal overreach is marked. Universities, Hospitals, Charity, and Literacy Planned Which institutions Christians founded, which had earlier parallels, how they changed over time, and what caused their wider spread. Women, Children, and the Protection of Life Planned Christian effects on infanticide, abortion, marriage, women's status, family law, and care during Roman epidemics. Would the World Be Better Without Christianity? Planned A counterfactual test of Christianity's effects that counts both institutions it built and harms committed in its name.

Christianity in History (0/5)

The church's institutions, conflicts, failures, and reforms across two millennia.

Does Religion Cause War? Planned War data, atheist regimes, religious identity, political causes, and whether faith predicts violence or prosocial behavior. Missions, Colonialism, and Institutional Abuse Planned Residential schools, cultural destruction, humanitarian work, clergy abuse, Rwanda, apartheid, and causal responsibility. Science and Christianity Planned The institutions and conflicts behind the claim that Christianity either created or obstructed modern science. The Crusades, Inquisition, and Witch Trials Planned What happened, how many people were harmed, which institutions were responsible, and how later retellings changed the record. The Early Church Planned Apostolic succession, persecution, councils, creeds, canon, and the diversity of the first Christian centuries.

Church History and Doctrine (0/6)

How Christian doctrines, institutions, movements, and divisions developed after the New Testament period.

Councils, Creeds, and the Trinity Planned What Nicaea and Chalcedon decided, how doctrine developed, and whether imperial power created Christian orthodoxy. Heresy and Early Christian Diversity Planned Ebionites, Marcion, Arius, and gnostic movements judged from their own texts and their opponents. Schism, Reformation, and Denominations Planned Why churches divided and whether persistent disagreement undercuts claims that Scripture is clear. The Apostolic Fathers and Early Christian Growth Planned What the earliest post-New Testament writers believed and how quickly Christianity spread through the Roman world. Theories of the Atonement Planned Ransom, satisfaction, moral influence, Christus Victor, and penal substitution compared against Scripture and moral objections. Women in Church Leadership Planned Junia, Phoebe, Priscilla, 1 Timothy 2, ordination, and evidence for women's roles in the earliest churches.

Psychology of Religious Belief (1/6)

Psychological explanations of belief tested for explanatory power and for whether they also apply to unbelief.

Ancient Monotheism Belief in one supreme creator is not a late invention: it is documented in Egypt by 1350 BC and in Babylon by 800 BC. But a recount of 1,291 societies kills the stronger claim. Foragers are the least likely people on earth to hold a high god, and 63% of the atlas's moralizing ones are simply Islam and Christianity. Evolutionary Debunking and the Accident of Birth Planned Whether evolved cognition or birthplace defeats religious belief without defeating other beliefs by the same reasoning. Motivated Reasoning on Both Sides Planned How identity, prior commitments, and social incentives distort apologetic and skeptical arguments alike. Projection Theories of Religion Planned Feuerbach, Marx, Freud, wish fulfillment, social control, and the evidence their theories predict. Religion, Mental Health, Education, and IQ Planned Measured correlations, confounders, subgroup differences, and what deconversion narratives can and cannot establish. Religion as a Cognitive Byproduct Planned Agency detection, intuitive purpose, and whether explaining how belief arises says anything about whether it is true.

The Church as Evidence (0/3)

Whether the behavior, endurance, growth, and decline of Christian communities count for or against their claims.

Do Converted Lives Change? Planned Addiction programs, crime, suicide, health, testimony selection, and what controlled outcome studies actually find. Global Christian Growth and Decline Planned The global South, fertility, conversion, retention, the rise of the nones, and uncertainty in demographic forecasts. Persecution and Martyrdom Planned Martyr counts, Open Doors methods, underground churches, failed communities, and whether endurance shows truth or commitment.