How Wrong Would We Have to Be About Old World Contact?

TL;DR

• To hide a significant Old World presence in pre-Columbian America, you need synchronized failure across archaeology, genetics, linguistics, and disease ecology.
• CRM archaeology alone has generated thousands of excavated sites and settlement sequences; a real Mediterranean or Near Eastern colony should show up everywhere in that dataset, not nowhere.¹
• Genome-wide data from hundreds of ancient individuals and many thousands of living Native Americans show overwhelmingly Northeast Asian ancestry with no pre-1492 West Eurasian component outside a few Polynesian-contact hints.²³
• No pre-contact Old World crops, livestock, writing systems, or disease patterns appear in stratified contexts, despite intensive work on mounds, middens, and paleoecology.[^hopewell]⁴
• A world where “Romans built the mounds” is not just “archaeologists were mean to a few stones”; it’s a world where four independent empirical programs have to be badly miscalibrated at once.

For broader context on pre-Columbian contact theories, see our survey of trans-oceanic contacts and comprehensive overview.

“The great tragedy of science — the slaying of a beautiful hypothesis by an ugly fact.”
— Thomas Huxley

1. From One Stone to Hidden Colonies#

Your previous piece asked: What if one of the canonical “hoax stones” turned out to be real? The answer was: archaeology eats crow, the “hoax” category gets reopened, and prehistory grows a weird new branch.

Here the bar is much higher.

We’re not just asking whether one Hebrew or Latin inscription could be genuine (as explored in our Tucson artifacts analysis). We’re asking what would have to break for substantial Old World presence—Romans in Arizona, Jews in Hopewell Ohio, Phoenicians in the Great Lakes, or “Old World mound builders”—to be real and leave almost no trace in four big datasets:

1. Archaeological deposits (artifacts, architecture, ecofacts).
2. Genomes (ancient and modern).
3. Languages and scripts.
4. Disease and domesticated species.

For each, we can ask three questions:

1. What should we see if there was sustained Old World contact?
2. What do we actually see?
3. Therefore, what would have to have gone wrong?

The answer, in short, is: a lot. Not a single mistake, but a stack of correlated failures.

2. Archaeology at Scale: Where Are the Romans in the Dirt?

2.1 How much ground have we actually covered?#

In 19th-century Ohio you could imagine that a handful of elite digs and a few spectacular mounds dominated the dataset. That world is gone.

Since the 1970s, cultural resource management (CRM) has become the dominant mode of archaeology in the United States. Roughly 80% of U.S. archaeologists work in CRM, and the majority of field projects are compliance-driven surveys and excavations tied to construction and infrastructure.¹ Annual CRM spending was already estimated in the hundreds of millions of dollars by the late 20th century; today it’s an industry-scale enterprise.¹⁵

A few concrete glimpses:

• One U.S. state, Louisiana, lists over 1,000 excavated sites in its state-level database alone.⁶
• The Digital Archaeological Record (tDAR) and similar repositories hold thousands of CRM reports, representing “some of the most significant CRM, Federal, and State archaeological work done in the United States in the last 50 years.”⁷
• Hopewell and Mississippian heartlands have been repeatedly excavated: classic work in the 19th century, large academic projects in the mid-20th, and multiple modern campaigns at sites like Mound City, Seip, and other Hopewell earthworks.⁸⁹

If the Eastern Woodlands’ mound builders were being significantly influenced by Mediterranean or Near Eastern colonists, the signal would have a lot of chances to show up—not just in one mound in Ohio, but in CRM trenches for highways, pipelines, subdivisions, and landfills over half a continent.

2.2 What a real colony should look like in the ground#

A sustained Old World presence—even a small one—doesn’t just produce inscriptions. It leaves a whole cultural and ecological package:

• Materials & technology
  • Old World alloys: lead-tin bronzes, particular ironworking signatures, recognizable slag.
  • Distinctive ceramics, glass, or bead styles.
  • Ship hardware, nails, spikes, ropework fragments, ballast stones.
• Architecture & layout
  • Forts, storehouses, or religious structures with Old World design elements.
  • Standardized measurements or construction modules that don’t match local patterns.
• Food & trash
  • Dietary shifts: wheat/barley/olive pits, grape pips, Old World domesticates in middens.
  • Butchered remains of sheep, goats, cattle, pigs, or chickens, with cut-mark styles familiar from Eurasian assemblages.

We know what this looks like because we do see it in genuinely colonial contexts: Norse Greenland and L’Anse aux Meadows, Viking Dublin, Phoenician colonies in the Mediterranean, Roman frontier forts, medieval trading enclaves in India and East Africa.¹⁰

At L’Anse aux Meadows, a tiny Norse outpost in Newfoundland, archaeologists identified turf buildings, iron nails, rivets, a bronze ring-pin, and domesticated animal bones consistent with Norse presence.¹¹ That’s the footprint of a marginal, probably short-lived waystation. A Roman-Jewish “Calalus” colony operating for generations in Arizona—or a Mediterranean elite shaping Hopewell mortuary practice—should be louder than that.

2.3 What we actually see#

In the mound areas and across North America, the record is stubbornly Indigenous:

• Earthworks and mounds with clear sequences of construction and reuse by ancestral Native groups.⁸⁹
• Lithic industries, ceramics, and copper working with deep local lineages and internal evolution, not sudden intrusive Mediterranean packages.⁹
• Diets based on local domesticates (maize, beans, squash, amaranth, chenopodium, sunflower) plus wild resources, not Near Eastern wheat-barley-sheep economies.⁴¹²

The Old World things that do show up in pre-Columbian contexts are precisely the ones we’d expect from Polynesian–South American contact:

• Sweet potato (a South American domesticate) is securely present in Polynesia before European arrival.¹³
• A chicken bone from Chile near Arauco, radiocarbon-dated to ~AD 1321–1407 and carrying a Polynesian mitochondrial haplotype, indicates Polynesian chickens reached South America.¹⁴

In other words: where there was contact, the dirt tells us so.

2.4 So what would have to have gone wrong?#

For “Old World mound builders” or “Romans in Arizona in real numbers” to be real yet invisible, several improbable things must stack:

1. Massive sampling bad luck

   • Thousands of CRM and research projects, many in the right places, somehow never intersect dense Old World deposits—or when they do, the finds conveniently sit in unexcavated patches.
   • This is possible for a handful of sites; it is very hard to square with continent-scale influence.

2. Systematic misclassification and discard

   • Old World metals, ceramics, or faunal remains are misidentified as modern contamination every time.
   • Distinctive alloys (e.g., Roman-style lead bronzes) are never subjected to compositional analysis that would flag them as anomalous.

3. Selective taphonomic annihilation

   • Old World organic traces decay faster than Indigenous materials in precisely those contexts where they could be diagnostic, while Indigenous materials survive to taunt us.
   • You would need a very perverse set of preservation rules.

4. Coordinated interpretive bias

   • Field archaeologists across decades and institutions independently decide that any anomalous Old World–looking item must be intrusive or fraudulent, and either discard it or bury it in grey literature that no one ever re-analyzes.

That’s not a sinister conspiracy so much as a cartoonishly strong prior that somehow never yields to contrary data. But once you factor in how prestige-generating a confirmed Roman, Jewish, or Phoenician colony would be (tenure-making, TED-talk-minting, textbook-rewriting), it starts to look less like mundane bias and more like a universe that’s actively trolling career incentives.

3. Genomes Without Ghost Romans#

If archaeology is “what’s in the ground,” genetics is “what’s in the bodies.” A sizable Old World presence has to go through actual people. Those people leave descendants or die out.

Either way, they leave DNA.

3.1 Modern Native American genetics: overwhelmingly Northeast Asian#

Genome-wide work since the early 2010s has produced a fairly stable picture:

• Native Americans largely descend from a “First American” population that diverged from East Asians ~25–36k years ago and spent time in or near Beringia before dispersing south.²¹⁵
• Additional streams from Northeast Asia complicated the picture (e.g., “Australasian” signals in some Amazonian groups, later Arctic gene flow), but these are all Asian, not West Eurasian.²¹²
• Studies using hundreds of thousands of SNPs in dozens of Native and Siberian groups find no widespread pre-Columbian West Eurasian component in mainland Native Americans; what West Eurasian ancestry modern Native people carry is overwhelmingly post-1492.²¹⁵

This is exactly what you’d expect if the main peopling was via Beringia, with later European and African admixture after 1492, plus a small Polynesian contact episode on the Pacific side.

3.2 Ancient DNA: time-slices across the Americas#

Ancient DNA (aDNA) lets you bypass modern admixture and see what populations looked like before Columbus.

• Moreno-Mayar et al. sequenced 15 ancient genomes from Alaska to Patagonia, showing branching and movements within a First American framework.³
• Posth et al. added 49 ancient individuals from Belize, Brazil, the Central Andes, and the Southern Cone, tracing multiple southward expansions and local turnovers—but again, all within an American–Asian genetic space.⁴
• Gnecchi-Ruscone et al. and others similarly find intricate mixtures of Native-American ancestry streams, not hidden Mediterranean or Near Eastern clusters.¹²

Crucially: these are genome-wide analyses using methods (PCA, ADMIXTURE, f-statistics, qpAdm) that are very good at picking up even modest West Eurasian admixture if it’s there in >~1–5% proportions.

Outside of:

• Norse Greenlanders (uncontroversial Europeans in Greenland), and
• Polynesian–Native admixture signals in some Pacific coastal/Island populations,

pre-Columbian West Eurasian ancestry is conspicuous by its absence.¹⁵¹⁶

3.3 What would significant Old World ancestry look like?#

Imagine three scenarios:

1. Big colony, many descendants.

   • A Roman-Jewish “Calalus” in Arizona that persists for centuries, intermarrying with local groups.
   • A Mediterranean or Near Eastern elite dominating a sizable slice of Hopewell or Mississippian society.

   After ~1,200+ years, you’d expect:

   • Local descendant populations with 5–20% West Eurasian ancestry.
   • Distinct Y-chromosome and mitochondrial haplogroups associated with the Mediterranean or Near East.
   • Ancestry clines around the colony site.

2. Small enclave, modest gene trickle.

   • A few hundred foreigners, limited mixing, but not zero.

   Expect:

   • At least a handful of individuals in regional time-series burials with measurable West Eurasian introgression.
   • Localized signals that show up in dense ancient sampling.

3. Total demographic failure: zero surviving descendants.

   • Colonists die of disease, famine, conflict, or sterility. No surviving children.

   Expect:

   • No genetic signal—but then you must explain why the colony left so little archaeological debris despite being large enough to build mounds or fight described wars.

Genetics doesn’t forbid a tiny, failed colony. It does make it extremely hard to reconcile large-scale or long-lived diffusionist scenarios with the current data.

3.4 So what would genetics have to get wrong?#

To smuggle in substantial Old World presence, you need another improbable pile-up:

1. Systematic contamination in the “wrong” direction

   • Ancient remains from potential contact zones (Ohio mounds, Southwest) would have to be systematically contaminated with Native rather than European DNA in ways that conveniently erase the signal, while contamination elsewhere produces recognizably European signals.
   • This is backwards from what lab contamination usually does.

2. Deep sampling blind spots

   • Old World–influenced groups just happen to be missing from both ancient and modern datasets. No one has sampled their descendants, and no representative burial has yielded DNA.
   • With dozens of ancient individuals and 52+ modern Native and 17 Siberian groups in Reich et al.’s classic work alone, plus later expansions,²¹² that’s possible but increasingly contrived.

3. Mis-modeled ancestry

   • West Eurasian admixture is being consistently misinterpreted as some other “ghost” Native or Asian lineage across multiple independent studies and labs.
   • Given how loudly West Eurasian ancestry stands out in global datasets, this requires a level of shared modeling pathology that starts to verge on cosmic prank.

You don’t need a literal conspiracy. You do need multiple teams in different countries, using different methods, all missing the same obvious pattern for a decade or more, in a field where finding such a pattern would be fame-making.

4. Languages, Scripts, and Cultural Memes#

Languages are exquisitely sensitive to contact. Even short-lived trade relationships can leave loanwords, especially for new technologies, foods, and prestige goods.

4.1 The one known long-distance contact: Polynesians#

The classic case:

• The Quechua/Aymara word for sweet potato kumar(a) closely matches Proto-Polynesian kumala, and the plant itself shows up in Polynesia pre-contact.¹⁶¹³
• This is so striking that some linguists call it “near proof” of Polynesian–Andean contact, even though we have no saga of the voyage.¹⁶

That is: a single crop and a single word are enough to convince cautious people that at least some canoe bumped into some coast.

4.2 What we don’t see for Mediterranean or Near Eastern contact#

Now ask: where are the Polynesian-style smoking guns for Romans, Jews, or Phoenicians?

If Mediterranean/Near Eastern traders or colonists were materially shaping mound-builder societies, we’d expect:

• Loanwords for:
  • Metals and metallurgy (iron, lead, bronze, forge, anvil).
  • Domestic animals (sheep, goat, pig, cow, horse).
  • Boats and navigation.
  • Religious ideas or titles (priest, temple, rabbi, bishop, etc.).
• Structural influence:
  • Number systems, calendrical terms, or writing practices.
  • Maybe even partial adoption of an alphabet or abjad, as happened along the Mediterranean coasts.

Instead:

• North American language families—Algonquian, Iroquoian, Siouan, Muskogean, Uto-Aztecan, etc.—show deep internal diversification and contact with each other, not clear borrowing from Latin, Greek, Phoenician, or Hebrew in core lexicon.
• There is no accepted pre-Columbian writing system in North America north of Mesoamerica; the only scripts we see are on the very stones currently treated as hoaxes.

Could linguists be missing subtle contact? Sure. Could they be missing widespread, structural impact from centuries-long Mediterranean colonies involved in mound-building? That’s harder. Linguists are very good at finding even faint Indo-European contamination in distant language families; if Hebrew or Latin had been camping out in the Ohio Valley long enough to teach anyone how to build earthworks, the odds that it left no trace in vocabulary look slim.

4.3 What has to go wrong in linguistics#

For diffusionist scenarios to be real but invisible linguistically:

1. All meaningful contact is non-linguistic

   • Colonists teach metallurgy, architecture, or religion, but somehow never exchange enough language for durable loanwords.
   • That’s not how social power usually works; prestige groups export both things and terms.

2. Loanwords exist but are systematically misanalyzed

   • Key loanwords are assumed to be internal developments or from other Native languages in every case.
   • Given the sharp phonological profiles of Semitic and Indo-European languages, this would require an impressive and oddly consistent blind spot.

3. Late linguistic replacement

   • Even if loanwords existed, massive later language shifts might have erased them. That’s plausible in some regions—but you’d need that erasure to be nearly complete and perfectly aligned with the archaeology and genetics also failing to see anything.

Again: not impossible, just increasingly rococo.

5. Disease, Domesticated Species, and the Missing Plagues#

If you drop Old World people into the Americas in meaningful numbers, you also drop their pathogens. That’s not optional.

5.1 The Columbian catastrophe#

Post-1492, Eurasian diseases—smallpox, measles, influenza, typhus, etc.—kill staggering proportions of Native populations, often 50–90% in epidemic waves.¹⁵ This implies:

• Near total immunological naïveté to those pathogens before European arrival.
• No earlier, large-scale exposure that would have provided partial immunity.

If sizeable Old World colonies had been operating in the Americas centuries earlier, trading and intermarrying, we’d expect:

• Earlier epidemic episodes recorded in oral history, archaeology (mass graves, disruption), or paleopathology.
• Some degree of pre-adaptation: not immunity, but perhaps less apocalyptic mortality in the 16th century.

Instead, disease patterns look like first contact on a continental scale.

5.2 Domesticated fauna and flora#

Old World colonists also bring:

• Animals: horses, cattle, sheep, goats, pigs, chickens.
• Crops: wheat, barley, rye, oats, olives, grapes, Old World pulses.

The archaeological record north of Mesoamerica shows:

• Dogs (Native), turkeys (in some regions), and later European livestock—after contact.
• No pre-Columbian sheep, goats, pigs, cattle, horses, or Eurasian cereals in secure contexts.¹⁶⁴

Where long-distance contact did happen—again, in the Polynesian–South American case—chickens and sweet potatoes did move, and we can see them.¹⁴¹³

So for Romans or Jews to be around long enough to shape mound-building but not long enough to leave a single secure pig bone in a pre-1492 layer, we need extremely constrained, ecologically sterile contact.

5.3 What ecological weirdness is required?#

To hide significant Old World presence ecologically:

1. Unusually “clean” colonies

   • Colonists somehow arrive without their usual livestock, or their animals all die before reproducing, leaving no trace.
   • Crops fail systematically, so there is no adoption of Eurasian cereals by Native farmers.

2. Pathogen bottleneck miracle

   • Colonists happen not to be carrying the big immunologically disruptive pathogens (smallpox, measles, etc.), or they arrive during rare disease-free windows.
   • Or their colonists are so few and so isolated that epidemics never take hold, despite enough contact to reshape monumental traditions.

3. Paleopathology misses mass mortality

   • Earlier epidemics happen but are archaeologically invisible, or are indistinguishable from other sources of stress.

You can arrange such a scenario in a novelist’s notebook—say, a single Roman shipwreck with a handful of survivors who breed local elites but happen to be between plague cycles and traveling unaccompanied by livestock. You cannot easily make it support widespread diffusionist claims about mound builders, Romans, and Jews on the landscape for centuries.

6. Incentives, Institutions, and the Alleged Conspiracy#

There are social dynamics here. Archaeologists dislike pseudoarchaeology in part because of its 19th-century ties to racist “lost white race” stories, and they train students to treat Bible-flavored stones with skepticism.¹⁷ That’s healthy.

But look at the incentive structure.

6.1 What a real Old World colony would buy you#

If you are the archaeologist who can prove:

• A Roman-Jewish colony in Arizona.
• A Hebrew-inscribed stone in secure Hopewell context.
• A Mediterranean-style trading post in the Great Lakes.

You get:

• A top-tier journal article.
• Near-instant celebrity in and beyond the field.
• Book deals, documentaries, and a likely permanent place in every “Prehistory” syllabus.

The idea that no one would take that trade because they’re too wedded to a “Beringia only” consensus is doing a lot of work.

6.2 What a conspiracy would actually require#

To maintain “significant Old World contact” as real yet invisible, you’d need some combination of:

• CRM firms, state archaeologists, academic researchers, and museum curators all independently suppressing anomalous finds across decades.
• Geneticists from multiple labs (Harvard, Copenhagen, Max Planck, etc.) declining to report West Eurasian ancestry in Native samples, or mislabeling it en masse.²³⁴¹²
• Linguists across many subfields ignoring clear loanword patterns.
• Paleopathologists politely failing to notice pre-Columbian Old World disease signatures.

And this would have to be maintained in an era of FOIA, leaks, preprints, and Twitter. If some CRM tech in Ohio really pulled a Roman gladius out of a Hopewell mound and was told to “lose it,” that story would be viral by the weekend.

Is there groupthink? Absolutely. Are there priors and moralized narratives? Of course. But the level of cross-domain silence needed to hide large-scale Old World settlement starts to look less like sociology and more like metaphysics.

7. Putting It Together: The Failure Stack#

Here’s the whole thing in one glance.

7.1 Reality vs. Hidden-Romans World#

Once you line the domains up, the question becomes less, “Could all these fields be wrong?” and more, “How many independent failure modes are you willing to stack before your prior shifts?”

8. So What Kind of Contact Is Still on the Table?#

None of this rules out small-scale, transient contact:

• We already accept Norse landings in Newfoundland.
• The Polynesian–South American connection is now backed by crops, bones, and genetics.¹⁴¹⁶
• It’s entirely plausible that other ships or canoes reached the Americas, traded a bit, and vanished.

Those are edge-of-the-map events. They don’t need to leave a thick archaeological or genomic trail. A single wreck or short-lived outpost can fall through every sieve we have.

But claims like:

• “Old Worlders built or decisively shaped the mounds,”
• “Romans or Jews were a significant presence in North America for centuries,” or
• “The Michigan Relics and Tucson crosses reflect a lost colonial history,”

are qualitatively different. They require scale and duration. Scale and duration create multiple, redundant traces. And redundant traces are exactly what our current systems are good at finding.

For specific examples of proposed Old World contact theories, see our articles on Phoenician claims and alternative histories.

So to live in a world where those heavy diffusionist scenarios are true, you have to live in a world where:

• Archaeology, genetics, linguistics, and ecology are all deeply mis-tuned in coordinated ways;
• The fields that successfully detected Norse Greenlanders and Polynesian chickens mysteriously fail to detect Old World actors who supposedly did much more.

You can choose that world, if you like your epistemology baroque. The more boring world—where Indigenous Americans built their own monuments, occasionally met strangers at the horizons, and then faced a catastrophic first contact in the 15th–16th centuries—has the advantage of being supported from four orthogonal directions at once.

The irony is that this “boring” world is already wild: multiple waves out of Beringia, lost lineages in Amazonia, canoe voyages across the Pacific, massive earthen geometries in Ohio visible from orbit. We don’t need Romans in Arizona to make it interesting. We’d just need Romans in Arizona to rewrite half of what we think we know about how evidence accumulates in the first place.

FAQ#

Q1. Does this mean pre-Columbian Old World contact is impossible?
A. No. Short, failed, or very small-scale contacts could easily leave little or no trace, just as some shipwrecks do today; what’s implausible is large, long-lived colonies shaping major Native traditions without showing up in multiple datasets.

Q2. Could unsampled descendant groups be hiding Old World ancestry?
A. There’s always room for surprises in unsampled or sparsely sampled communities, but the broad patterns—continental-scale First American ancestry with no obvious pre-1492 West Eurasian component—are hard to overturn with one or two outliers.

Q3. What about claims of Old World artifacts in private collections or one-off digs?
A. Isolated finds with poor context are exactly where hoaxes and misidentifications thrive; without secure stratigraphy, dating, and replication, they can’t outweigh the convergent negative evidence from better-documented sites.

Q4. How is this different from Polynesian contact with South America?
A. Polynesian contact is supported by converging evidence—sweet potatoes in Polynesia, a pre-Columbian chicken bone in Chile with Polynesian DNA, and genetic signals in some populations—precisely the kind of multi-domain corroboration missing from Mediterranean diffusionist claims.

Q5. If one canonical “hoax stone” turned out to be genuine, would all this crumble?
A. It would force a recalibration and a re-audit of similar artifacts, but unless that stone pointed to a large, enduring colony with independent support from genetics and ecology, the broader arguments here would still constrain how big that contact episode could have been.

Footnotes#

Sources#

1. Green, William, and John F. Doershuk. “Cultural Resource Management and American Archaeology.” Journal of Archaeological Research 6, no. 2 (1998): 121–165.
2. “The development of Cultural Resource Management in the United States.” In Archaeological Research, Preservation Planning, and Public Education in the Northeastern United States.
3. Louisiana Division of Archaeology. “Excavated Sites Database.”
4. The Digital Archaeological Record (tDAR). “CRM Archaeology and tDAR” collection.
5. National Geographic. “Inside the secret world of the Hopewell Mounds—our newest World Heritage site.” November 20, 2023.
6. Hopewell Archeology Newsletter, various issues detailing excavations at Mound City and related sites.
7. Reich, David, et al. “Reconstructing Native American Population History.” Nature 488 (2012): 370–374.
8. Moreno-Mayar, J. Víctor, et al. “Early human dispersals within the Americas.” Science 362, no. 6419 (2018): eaav2621.
9. Posth, Cosimo, et al. “Reconstructing the Deep Population History of Central and South America.” Cell 175, no. 5 (2018): 1185–1197.
10. Gnecchi-Ruscone, Guido A., et al. “Dissecting the Pre-Columbian Genomic Ancestry of Native Americans.” Science Advances 5, no. 12 (2019): eaaw3019.
11. “Genetic history of the Indigenous peoples of the Americas.” Wikipedia, overview article.
12. Storey, A. A., et al. “Radiocarbon and DNA evidence for a pre-Columbian introduction of Polynesian chickens to Chile.” PNAS 104, no. 25 (2007): 10335–10339.
13. “Pre-Columbian transoceanic contact theories.” Wikipedia, summary of contact debates.
14. Feder, Kenneth L. Encyclopedia of Dubious Archaeology: From Atlantis to the Walam Olum. Greenwood, 2010.
15. “Pre-Columbian Trans-Oceanic Contact – How Humans Evolved.” Snakecult.net, 2025.
16. “If the Stones Were True: How One Proven ‘Hoax’ Would Rewrite American Prehistory.” Snakecult.net, 2025.