More Options

Pathology or Paradigm Shift? Human Evolution, Ad Hominem Science and the Anomolous Hobbits of Flores


Kenneth W. Krause

Skeptical Inquirer Volume 33.4, July / August 2009

Every once in a while, there comes to light a fossil that shakes the foundation of paleoanthropology to its very core and forces us to reconsider what we thought we knew about human evolution.

—Donald C. Johanson, Lucy’s Legacy

Wonderful and sometimes dangerous things can come in very small packages. The discovery in September 2003 of an extremely petite, human-like creature was certainly wonderful. But what could be so dangerous about a long-dead hominin little more than a meter tall with a brain the size of a newborn baby’s? The answer reveals a deep and still widening rift in the scientific community between one faction standing guard over received wisdom and another bloc poised to make scientific history.

Dubbed Homo floresiensis by their co-discoverers, these “Hobbits” evidently thrived for thousands of years beneath the capacious limestone dome of the Liang Bua (LB) cave on the East Indonesian island of Flores. In sediments approximately 18,000 years old, archaeologists unearthed a nearly complete skeleton, including the skull. LB1, as she is known, was an adult female—about thirty years old judging from tooth-wear, but her cranium was considerably smaller than that of any other known hominin from the genus Homo. While modern human brains average 1,350 cubic centimeters (cc) in volume and H. erectus brains averaged 900 cc, LB1’s amounted to just 417 cc.

Yet there were signs that these small-brained Hobbits were no dummies. Along with the skull and bones, stone tools and charred animal remains were recovered, suggesting that the LB hominins had maintained an advanced hunting culture. Until that point, most experts had insisted that such a culture was inconsistent with tiny brains and, perhaps, with any species of hominin except H. sapiens. As news of the Hobbits spread, challengers proposed a number of intriguing theories to explain why these creatures should not be classified as a new species. Their opponents, however, joined by LB1’s co-discoverers, have not relented. The debate over H. floresiensis has in fact intensified since 2003, becoming ever more contentious and, at times, downright bitter.

The Early Nature Papers: Building a Case for Homo floresiensis

Homo floresiensis challenges us because she is so unexpected, because she does not fit with many preconceptions about how humans evolved and behaved, and what they should look like.

—Michael Morwood, The Discovery of the Hobbit

A 3D cast of a skeleton of the controversial meter-tall adult hominin named Homo floresiensis by its discoverers and popularly called the Hobbit.

A 3D cast of a skeleton of the controversial meter-tall adult hominin named Homo floresiensis by its discoverers and popularly called the Hobbit.

Photo: Tim Wiencis/ Splash News

In October 2004, LB1 was introduced to the world in a pair of articles in Nature headlined by two Australians, paleoanthropologist Peter Brown and archaeologist Michael J. Morwood. From the beginning, Brown emphasized the specimen’s “unique mosaic of primitive and derived traits” (Brown 2004). LB1’s diminutive body size generally suggested some association with Pliocene australopithecines, Brown judged, yet her facial and dental traits tended to link her more closely to larger Homo specimens from the Pleistocene.

To account for the Flores hominin’s small stature, Brown invoked the “island rule,” or the selective advantage of insular dwarfing in the context of isolated, predator-free environments marked by reduced competition and resources. Smaller species would be favored in such situations due simply to their reduced energy requirements. H. floresiensis, Brown concluded, resulted from dwarfing either on Flores itself or in another insular southeast Asian environment prior to emigration.

From the outset, both teams considered the possibility that LB1 was a unique, perhaps diseased, individual. Morwood promptly rejected that hypothesis, however. Referring to a mandibular left third premolar (from LB2) found in significantly older deposits, he argued that they were “not dealing with an abnormal individual but a long-standing population” (Morwood 2004). Morwood’s team also identified many of the charred faunal remains, some of which belonged to Komodo Dragons and a dwarfed species of Stegadon—a now-extinct group of animals related to mammoths, mastodons, and elephants. In the same location, they found “big game” stone artifacts including “points, perforators, blades, and microblades” that the team inferred “were probably hafted as barbs” by the Hobbits. Because of these artifacts’ close proximity to other tools likely crafted by H. erectus, and because of LB1’s morphological consistencies with H. erectus, Morwood initially judged H. floresiensis to have descended from that species.

More Hobbit remains were excavated from the LB cave in 2004, including the right humerus and right ulna of LB1, a second adult mandible, and various postcranial features from other individuals. In a third paper published in October of the following year, Brown and Morwood announced the discovery of at least seven more inhabitants (LB3–LB9) that had been recovered from sediments dating between 95,000 or 74,000 to around 12,000 years ago (Morwood 2005). All were diminutive—probably even smaller than LB1. On the one hand, this new evidence solidified the team’s rejection of the pathology hypothesis; on the other, it compelled them to reconsider H. erectus as the Hobbits’ likely ancestor.

Initially, Morwood and Brown noted that the second, chinless mandible (LB6)—though having smaller teeth than those belonging to LB1—was quite similar to LB1’s mandible in general size, morphology (form and structure), and symphysis (where two halves of the mandible unite). Importantly, LB6 was also estimated to be about 3,000 years younger than LB1, thus providing “another line of evidence,” the authors argued, “that LB1 was not aberrant, but is instead representative of a long-term, morphologically unique, small bodied population.”

The team also determined that the LB1 and LB6 mandibles were more like early east African or Georgian H. erectus than either Javan or Chinese H. erectus, and that their dental symphyses most resembled even earlier hominins, including Australopithicus afarensis.

Morwood and Brown then examined the new arm and leg bones, first comparing Hobbit midshaft circumferences to those of modern African Pygmies. The width of LB8’s tibia, for example, rested within the Pan (chimpanzee/bonobo) and Pongo (orangutan) range but not within the Homo range. Similarly, LB1’s humerus and ulna were significantly thicker than those of the Pygmies.

Second, they compared the Hobbits’ relative limb and body proportions to those of various apes and early hominins. With a relatively long arm-to-leg ratio and an antero-laterally flared pelvis, the team found that LB1’s humerofemoral index was outside the range of variation for H. sapiens but identical to A. afarensis and that her body proportions were different from all hominins—including H. erectus—except A. afarensis. “Abnormal growth seems an unlikely explanation,” the team concluded, “as growth-hormone-related dwarfism and microcephaly [literally, small brain] in modern humans result in normal limb and pelvic proportions.” Although adamant that the LB hominins were not H. erectus or H. sapiens, Morwood and Brown left open the question of whether they had descended from an even earlier species of Homo or from an australopithecine.

The Early Detractors: Microcephaly Reconsidered

The twist is that there was a rush to judgment, because the temptation to be recognized as discoverers of a new human species was too great to resist.

—Maciej Henneberg, The Hobbit Trap

An artist’s impression of Homo floresiensis

An artist’s impression of Homo floresiensis.

Photo: Peter Schouten/AFP/Getty Images

Shortly after the announcement of H. floresiensis, Polish-born biological anthropologist Marciej Henneberg and Australian paleoanthropologist Alan Thorne published their criticisms of Brown and Morwood’s conclusions in a non-peer-reviewed context, briefly proposing a “simple explanation” for the Hobbits’ odd features (Henneberg and Thorne 2004). Secondary microcephaly (secondary, meaning occurring later in development), they argued, could explain LB1’s paradoxically small braincase (five to six standard deviations below the modern average) relative to her “normal” face, nose, and jaw (three standard deviations below average). Comparing two microcephalic skulls described in the archaeological records to the skull of LB1, the authors found that not one of the fifteen dimensions evaluated differed by more than 2.5 standard deviations. Henneberg and Thorne also described LB1’s orthodontic crowding and rotation problems and her receding chin as consistent with the suggested growth disorder.

In their response, Brown and Morwood acidly described Henneberg and Thorne’s paper as “an extremely poorly informed, and ill designed, piece of ‘research’ [that] could not have been published in a substantial peer reviewed journal” (Brown and Morwood 2004). Henneberg later calculated the likelihood that LB1’s cranial capacity could evolve according to standard gradualist theory and consistent with the documented hominin fossil record. He came up with a probability of “much less than 0.0001” (Henneberg 2007). Morwood crustily dismissed that claim too. Given the wealth of evidence in favor H. floresiensis, he said, “the probability of Henneberg’s claims having any substance . . . is much less than 0.0001.”

The first highly detailed dissenting analysis, however, was led by Teuku Jacob, Indonesia’s then-premier paleoanthropologist, sadly now deceased, who questioned why supporters of the H. floresiensis taxon had compared LB1 mostly with humans from other parts of the world. Jacob offered and tested an alternative hypothesis—that “LB1 was an Australomelanesian H. sapiens who manifested microcephaly, which is commonly accompanied by other developmental abnormalities” (Jacob 2006).

Discussing specific skeletal features first, Jacob contended that none of the ninety-four previously described cranial features of LB1 or the forty-six features of the two mandibles were outside the range for regional modern humans. Similarly, with respect to the jaws’ supposedly most distinguishing feature—absence of a chin—Jacob referred to the Rampassa pygmies currently living near the Hobbits’ cave, 93 percent of whom display flat or even negative chins.

Next, Jacob turned to the issue of pathology. Microcephaly, he pointed out, is a clinically heterogenous condition potentially caused by various genetic, chromosomal, and environmental problems affecting prenatal and postnatal development. It can also result in numerous physical symptoms in addition to small brain size, including craniofacial and postcranial asymmetries. Indeed, Jacob found LB1’s face to be highly asymmetrical: six of seven measured areas of its right side were as much as 40 percent larger than those on the left.

Also, he discovered that LB1’s long bones were not simply “robust,” as Morwood and Brown had described. Rather, CT scans revealed abnormally thin cortical bone and very large marrow cavities indicative, according to Jacob, of weak muscle attachment markings and abnormal growth. In the end, Jacob’s team concluded that, contrary to Brown’s assessment, LB1’s traits were “not primitive but instead regional ... and not derived but strikingly disordered developmentally.”

Two additional critiques were published later that year. First, Robert Martin, an allometry (relative growth) specialist at the Field Museum of Natural History in Chicago, refuted Morwood’s initial claim that the Hobbits had descended from H. erectus (Martin 2006a). LB1’s braincase was smaller than any other known hominin less than 3.5 million years old, he observed, and much too small to derive from H. erectus by normal dwarfing.

However, a pair of British paleontologists, Eleanor Weston and Adrian Lister, have recently discovered that the brains of certain extinct island hippos had shrunk to a size 30 percent smaller than would otherwise be predicted under the traditional dwarfing model (Weston and Lister 2009).

Agreeing generally with Jacob’s diagnosis of microcephaly, Martin remained unbothered by the fact of multiple LB specimens (as many as sixteen, according to Morwood’s latest estimate). First, there was only one skull, that of LB1. Second, although the discovery of a second chinless mandible raised questions (if chinlessness is interpreted as a sign of microcephaly), the dating of the second mandible was itself dubious, Martin argued, because there was strong evidence—“the mingling of at least two different assemblages of stone tools” in particular—that the dated cave sediments had been disturbed. He also denied that the Hobbits had actually produced the tools associated with them.

Martin conceded an important point, however. LB1’s forelimb/hindlimb ratio did resemble that of something more primitive than H. erectus. It was “marginally possible,” therefore, that LB1’s remains provided “evidence of a new species from a lineage that diverged at a very early australopithecine stage, about 3 mya [million years ago], when cranial capacity was still very small.” Nevertheless, he cautioned, the existing fossil record simply did not support the hypothesis that so many more modern Homo traits could convergently evolve into the proposed H. floresiensis lineage.

Then, during a brief review of the contemporary Hobbit literature, University of California at Berkeley biologist Gary Richards emphasized a similar point (Richards 2006). Richards first proposed a genetic rather than a pathological cause of the Hobbits’ morphology—a mutation in the MCPH gene family combined with a modification of the growth hormone/insulin-like growth factor I axis). But he also challenged the new species proponents to account for Brown and Morwood’s novel evolutionary scenario. The Hobbits’ phylogenetic link “to any taxon other than H. sapiens,” he warned, “results in a problem wherein the similarities between H. floresiensis and H. sapiens would have to arise by parallel evolution at the same time that the new species is undergoing significant genetic drift or selection for a reduction in stature and brain size.”

Rejoinder: The Brain of H. floresiensis

If this is the best evidence that can be produced ...we suggest that the authors reconsider their position on the microcephalic hypothesis.

—Dean Falk (2005a)

Clearly, LB1’s tiny brain had seized center stage in the hotly contested Hobbit debate. Experts can’t study the brains of long-dead specimens directly, of course, but they can fashion endocast molds of braincase interiors. Although LB1’s cranium was far too fragile to render a common plaster cast, Brown fashioned virtual endocasts using CT scans, resin, and a rapid-prototyping process called stereolithography.

Florida State University’s Dean Falk, a leading expert on hominid brain evolution, decided to compare LB1’s virtual endocasts with molds made from the braincases of great apes, an australopithecine, an H. erectus, an average-sized H. sapiens, a pygmy, and a microcephalic H. sapiens (Falk 2005b). She found that although LB1 closely resembled A. africanus in terms of relative brain-to-body size, its brain’s general shape was most similar to that of H. erectus. Importantly, Falk observed as well that the Flores hominin’s endocast bore little likeness to that of the pygmy and least of all to the microcephalic.

Falk’s most remarkable discovery, however, was that LB1 had possessed a “well-convoluted” brain that could not have simply been a scaled-down version of an H. sapiens or an H. erectus brain. Referring to the Hobbits’ big-game stone technology and evident use of fire and cooking, Falk spotlighted LB1’s extremely wide temporal lobes and expanded frontal polar region, the latter of which consists of Brodmann’s area 10—probably involved in planning and initiative taking—in humans. In the end, Falk’s team (including Brown and Morwood) settled on two potential evolutionary scenarios: H. floresiensis either dwarfed under the island’s unusual allometric constraints or shared with H. erectus an unknown, small-bodied, and tiny-brained ancestor.

But, of course, Falk’s claims did not go unchallenged. German neuroscientist Jochen Weber, for example, analyzed nineteen different microcephalics (with a mean brain capacity of 404 cc) and found that seven, like LB1, presented an enlarged Brodmann’s area 10 (Weber 2006). Robert Martin argued that LB1’s tiny brain could not possibly result from the normal, non-pathological dwarfing of H. erectus and that the stone tools recovered from the LB cave were of a type never to have been associated with any species other than H. sapiens (Martin 2006b).

Australian archaeologist Adam Brumm, however, disagreed with Martin on the second point. Brumm’s group compared the LB tools to some 500 stone artifacts excavated from the Mata Menge site in the Soa Basin of central Flores (just fifty kilometers west of the LB cave) dated from 840,000 to 700,000 years ago—at least 500,000 years prior to H. sapiens (Brumm 2006). Both assemblages evidenced the same use of raw materials and a very similar freehand reduction technique. Both sites, in fact, produced the same types of tools of similar maximum dimensions. According to Brumm, “the stone artifacts from Mata Menge and Liang Bua represent a technology made by the same hominin lineage. Pronouncements that H. floresiensis lacked the brain size necessary to make stone artifacts,” he concluded, “are therefore based on preconceptions rather than actual evidence.”

Professor Falk eventually published two additional papers on the ever-enigmatic brain of LB1. In the first study, she compared LB1’s virtual endocast to those of nine “heterogenous” microcephalics and ten normal humans (Falk 2007). She concluded that LB1’s unique features should not be attributed to pathology because (1) “the frontal breadth relative to cerebellar width and lack of cerebellar protrusion” classified LB1 at “100% probability with normal H. sapiens rather than microcephalics” and (2) as she had already noted, LB1’s brain shared too many crucial affinities with those of H. erectus and A. africanus.

In the second paper, Falk identified seven distinct features of LB1’s brain that were derived and not pathological, declaring that “to date, we are unaware of descriptions in the literature of microcephalic brains ... that manifest anything like the suite of derived cortical features seen in LB1’s virtual endocast” (Falk 2009a). But perhaps even more interestingly—and refreshingly—Falk seized this opportunity to expound on what she deems a profound “shift in the big picture” of hominin brain evolution. Conventional wisdom has long demanded that the evolution of intelligence was the product of brain size alone. However, a Neanderthal- or H. sapiens-sized brain might not be necessary to the task. Hominin brain evolution—and, thus, selection for intelligence—might very well have proceeded in some species, like H. floresiensis, through cortical reorganization as well.

Mounting Evidence: The Body of H. floresiensis

Now that much of the fuss has died down, and the validity of the new species has been established apparently to the satisfaction of most paleoanthropologists, it has become possible to discuss its affinities soberly.

—Colin Groves (2007)

Three teams have published general studies of LB1’s cranial morphology in recent years, and each of the three arrived at a similar conclusion. An Australian group, led by anthropologist Debbie Argue, decided that LB1’s cranium did not resemble those of pygmies and was unlikely to belong to a microcephalic H. sapiens (Argue 2006). Instead, she proposed that LB1’s skull was most similar to that of H. ergaster, an African species from Koobi Fora sometimes classified as H. erectus, and that its limb proportions most resembled those of A. garhi. Then, after conducting the first scaling analysis of LB1’s cranium, American paleobiologist Adam Gordon found that in the absence of scaling, LB1’s skull was most similar to that of non-Asian H. erectus and H. ergaster and that when modern human skulls were scaled to LB1’s size, the Flores hominin’s cranium proved “even more distinct” from H. sapiens (Gordon 2008). Finally, a second American team led by anatomist Karen Baab agreed generally that the shape of LB1’s skull did not resemble that of small modern humans (Baab and McNulty 2008). They also addressed Jacob’s concern regarding LB1’s facial asymmetry, finding LB1’s features to be “consistent with the degree of asymmetry found in extant apes and humans,” and, in fact, “less asymmetrical than some other fossil Homo crania.”

Studies dedicated to specific elements of the postcranial remains bolstered the new species hypothesis as well. For example, Susan Larson, anatomist at the Stony Brook University School of Medicine, analyzed the shoulder bones of two LB specimens and the “Nariokotome” boy (an early H. erectus) and compared them to the shoulders of modern humans (Larson 2007). In both the Hobbits and H. erectus, Larson discovered an “unexpected combination of primitive and derived characteristics” in low humeral torsion, a relatively short clavicle, and a more modern scapula.

But it was the feature of low humeral torsion (technically, “the orientation of the humeral head relative to the mediolateral axis of the distal humerus”) that she found most remarkable. By contrast, high torsion characterizes modern humans. African apes display it too, as Larson explains, but it was subsequently lost at some unknown point in early hominin evolution and finally recovered by later Homo. In any case, for Larson, reduced torsion in both Nariokotome and the Flores hominins—distinct creatures from very different times and places—reveals a “transitional stage” in pectoral girdle evolution and strong evidence that the Hobbits were not H. sapiens, normal or pathological.

At about the same time, Matthew Tocheri, an expert in wrist evolution at the Smithsonian Institution’s National Museum of Natural History, published his team’s analysis of three of LB1’s wrist bones—the trapezoid, scaphoid, and capitate (Tocheri 2007). Individually, each bone revealed an important difference between the wrists of LB1 and modern humans. Unlike the human’s boot-shaped trapezoid, for example, LB1’s trapezoid is more wedge-shaped, like those of other primates. Similarly, the Hobbit’s capitate lacks the larger palmarly placed articular surface for the trapezoid common to both modern humans and Neanderthals and instead displays the “waisted neck” feature found in apes and australopithecines. Finally, LB1’s scaphoid shows an articular surface for the trapezium that does not reach out onto the scaphoid tubercle, as in humans and Neanderthals. According to Tocheri, these differences were not likely caused by developmental abnormalities like microcephaly because the shapes of these bones are determined well before the time when growth rate genes express themselves.

More generally, Tocheri discovered a significant functional distinction between LB1’s hand and the hands of other modern humans. Humans, Neanderthals, and the 800,000-year-old H. antecessor all possessed a “complex of five bones that mesh together to ease stress on the wrist when the hand is used forcefully, for example, in pounding large tools or in precision work” (Gibbons 2007). LB1, whose hand resembles more ancient hominins like the 1.7-million-year-old H. habilis, lacks this bone complex.

For Tocheri, LB1’s unique wrist morphology supports the hypothesis that the Hobbits descended from a hominin that migrated out of Africa prior to the evolution of the modern wrist.

Stony Brook University’s William Jungers arrived at a similar conclusion in his group’s new paper in May discussing the Hobbits’ unusually large feet, measuring 196 mm in length (Jungers 2009b). LB1’s foot-to-femur ratio was about 0.7, in fact, which “far exceeds the upper limits for modern humans,” Jungers said, “and instead overlaps with bonobos.” Indeed, many Hobbit tarsal bones display characteristics conforming to ape but not human samples. According to Jungers, LB1’s overwhelming number and variety of primitive traits—in its feet and throughout its cranial and postcranial skeleton—proves that the Hobbits were not pathological H. sapiens but rather descendents of a “primitive hominin that established a presence in Asia either alongside or at a different time than H. erectus.”

The Latest Criticisms and Responses: Pygmies, Laron Syndrome, and ME Cretins

Critics were unswayed, saying that even if one kind of pathology has been refuted, hundreds of others remain possible.

—Elizabeth Culotta (2008a)

Although many detractors have never withdrawn their argument that Hobbits were simply microcephalic H. sapiens, previously uninvolved authors have recently joined what Henneberg once labeled the “pathology group” and proposed fascinating new explanations for the Flores hominins’ peculiar physiology. For example, South African paleontologist Lee Berger questioned the new species hypothesis by comparing the LB specimens to a sample of twenty-five pygmy humans excavated in 2006 and 2007 from ten burial caves among the rock islands of Palau, Micronesia (Berger 2008). Radiocarbon dated to about 1,000 to 3,000 years old, the Palauan remains are “small,” according to Berger, “even relative to other pygmy populations.” Indeed, they “approximate in size H. floresiensis specimens and small members of the genus Australopithecus.”

Aping Peter Brown’s original characterization of LB1, Berger also noted that the Palauans displayed “some morphological features that are primitive” (dentition, jaw, and skull details) and “craniofacial traits that are considered to be uniquely derived ... in H. sapiens” (a reduced chin, large teeth, small orbits, and a ridge bone at the eyebrow). That the Hobbits and Palauans shared some morphological features, Berger argues, should “caution” us that “care must be exercised in interpreting their taxonomic and phylogenetic significance.” The Palauan pygmies, he adds, exemplify “the regularity with which small body size—physiological dwarfing—emerges in island contexts.” Falling back on a more familiar claim, Berger attributed LB1’s exceptionally small brain—which was not found among the Palauans—to “congenital abnormalities.”

But Scott Fitzpatrick, an archaeologist from North Carolina State University in Raleigh, has cast “serious doubts” on what he calls Berger’s “fundamentally flawed” claims and methods (Fitzpatrick 2008). He denies that Berger’s sample represents a true population or, in fact, anything other than an isolated clan. Fitzpatrick’s decade of work in Palau suggests, to the contrary, that early Palauans were normal sized and could not have undergone insular dwarfing—in part because Palau, unlike Flores, has never been isolated. Moreover, Australian anatomist Andrew Gallagher agreed with Fitzpatrick, concluding that the circumstances surrounding Berger’s Palauan sample were “likely unconnected to the prolonged genetic isolation envisaged for Middle Pleistocene Flores” (Gallagher 2008).

Consistent with Richards’ previous claim that LB1 had suffered from a defect along the GH/IGF-I axis, an Israeli team led by Israel Hershkovitz offered yet another genetic explanation for the persisting Hobbit conundrum: that “LB1 is not a new Homo species but a local variant of LS [Laron Syndrome]” (Hershkovitz 2007). LS, or primary growth hormone insensitivity, is a recessively inherited malady resulting from deletions or mutations within the growth hormone receptor (GH-R) gene that causes reduced growth hormone signal transmission. “The resulting phenotype,” Hershkovitz observes, “is extremely low stature and small head, but normally shaped bones.”

After comparing the Hobbits to at least seventy-nine LS patients and related literature, Hershkovitz isolated thirty-three shared skeletal morphologies that he deemed significant, including overall stature (106 cm for LB1 and 95–136 cm for LS patients) and skull size (5.5 standard deviations below the norm for LB1 and “up to” 5.0 SD below the norm for LS patients). With respect to brain size, Hershkovitz confessed that LS patients have smaller than normal crania, “although not to the same extent as LB1.” But to account for the difference, he vaguely noted “a high degree of association between microcephaly and growth failure in general.” Even so, the author felt it necessary to caution that “one should not expect complete cranial morphological similarity” because of the “numerous GH-R mutations involved in LS” and because his Mediterranean sample of LS patients would certainly differ from their South Asian and Pacific counterparts.

In closing, Hershkovitz referred (as Berger would) to Brown’s famous allusion to LB1’s combination of primitive and derived characteristics as the best argument in favor of distinguishing Hobbits from H. sapiens. But surely “nobody would argue,” Hershkovitz correctly (though somewhat sarcastically) noted, that “LS patients who also manifest a similar combination ... are direct descendents of Homo erectus ... or of australopithecines.”

Dean Falk and her international team of scientists, however, have very recently referred to the LS argument—and all other pathology hypotheses—as unscientific “cognitive dissonance” (Falk 2009b). Falk began her assault on Hershkovitz’s paper by identifying the ten skeletal features (in addition to overall short stature) that have “traditionally” distinguished LS patients from nonaffected others. Six of those ten features were not included in the thirty-three traits listed by Hershkovitz in 2007. Moreover, Falk continued, two of Hershkovitz’s thirty-three traits directly contradicted the traditional literature. Falk implied, in other words, that many of Hershkovitz’s criteria were completely alien to the LS diagnostic standards.

Nevertheless, she subjected LB1 to not only the traditional criteria but to at least eleven of Hershkovitz’s other criteria as well and found that “except for short stature and nondiagnostic angles of the femur, LB1 looks nothing like patients with LS.” Nor was Falk impressed by Hershkovitz’s cautionary caveat that “one should not expect complete cranial morphological similarity.” “Despite all the variation in LS,” she rebuked, “Hershkovitz et al. have [neither] described nor illuminated even one patient with LS who looks anything like LB1.”

In an effort to explain the substantial number of tiny Hobbits (and their inferred intelligence), Peter Obendorf of RMIT University in Melbourne, Australia, proposed an environmental rather than a strictly genetic explanation: myxoedematous endemic (ME) cretinism (Obendorf 2008). Suffering from a lack of iodine, ME cretins are born without a functioning thyroid. The congenital hypothyroidism that results can lead to “severe dwarfism and reduced brain size,” according to Obendorf, “but less severe mental retardation and motor disability than in neurological endemic cretinism.”

Obendorf’s team compared previously published descriptions of cretin physiology with that of LB1 and LB6 and claimed to have distinguished a substantial number of common features capable of confirming their hypothesis. Although they were “unaware of any recent ME cretinism on Flores,” the Australians did offer possible scenarios in which the Hobbits may have been subjected to three environmental factors—low iodine, low selenium, and high thiocyanate—the combination of which “induces thyroid necrosis in rats, a model for ME cretinism.”

Their arguments have not gone uncontested, however. In a new and quite brusque retort, Jungers (along with Falk, Tocheri, Larson, and Morwood, among others) contended that Obendorf’s cretinism hypothesis “can be rejected due to numerous errors of fact and unsubstantiated speculations” (Jungers 2009). Indeed, the two teams’ assessments of the facts are so disparate, one wonders whether they even considered the same data. Where Obendorf argues, “Human cretins have long arms relative to legs,” Jungers counters, “The humerofemoral index of cretins is normal for humans.” Where Obendorf claims, “Frontal sinuses are absent in ... LB1,” Jungers replies, “LB1 exhibits well-developed sinuses.” The former says that the cretin clavicle is “consistent with that of the LB1 clavicle.” But the latter argues that “the low claviculohumeral index of LB1 is outside the known range for cretins.”

But there was one significant point of agreement between the two parties. The most crucial feature of pertinence is the size of each sample’s pituitary fossa (a notch in the skull where the pituitary gland rests). Even so, while Obendorf claims that LB1’s fossa was “enlarged ... and a clear sign of pathology,” Jungers argues that “the pituitary fossa of LB1 is damaged, but not enlarged.” The scientific consensus on this point, however, seems to lean toward Jungers (Culotta 2008b). And although Obendorf believes that “numerous features” support the ME cretin hypothesis, even he refers to it as categorically “tentative.”

All hypotheses positing H. sapiens Hobbits, diseased or not, must contend with an archaeological record that, according to Dutch researcher G.D. van den Bergh, clearly distinguishes the periods during which each hominin occupied the LB area (van den Bergh 2008). The putative H. floresiensis era is well evidenced by Stegodon remains, for example, and was immediately preceded by major volcanic activity marked by thick deposits of volcanic tuff. But “these two endemic species,” van den Bergh observes, “are absent from deposits overlying white tuffaceous silts where the first skeletal and behavioral evidence for modern humans occurs” about 11,000 years ago.

Henneberg, for one, has expressed doubts about the accuracy of LB sediment dating, arguing that the deposits “could have been disturbed by flood actions.” He insists that experts should radiocarbon date LB1’s remains directly, a process he has repeatedly offered to fund personally. Even so, Henneberg has yet to publish a detailed criticism of the accepted chronology.

Conclusion: Persisting Issues and the Future of H. floresiensis

Most probably, we are on the threshold of a profound transformation of our understanding of early hominid evolution.

—Robin Dennell and Wil Roebroeks (2005)

Like Johanson’s famous australopithecine Lucy, the anomalous Hobbits of Flores may soon transform our understanding of human evolution forever—perhaps even revolutionize the very definition of humanity. But, regrettably, the debate—and, thus, the course and character of science itself—has been hideously defaced by unprofessional jealousy, rancor, and ad hominem attacks.

Morwood has likened detractors to flat-earthers, for example, while Robert Eckhard, a distinguished member of Jacob’s team in 2006, has averred a “racist” effect to the new species model. Indeed, individuals on each side have accused their counterparts of not being “real scientists.” If not their personal ethics, then their professional duties to things larger than themselves—the institutions of dispassionate scientific research and public education in particular—should suffice to prevent these kinds of embarrassing sideshows from occurring or escalating in the future.

The pathology debate, according to Jungers, is “officially over.” Perhaps. But detractors continue to raise important questions. Why, for instance, has only one skull been found if the species lived on Flores for 70,000 years? Should the textbooks be rewritten based on that single cranium? Is it not peculiar that we have discovered only one tiny-brained species capable of using tools and that it was located only on the remote island of Flores? And although it’s true that tropical environments are less than conducive to molecular preservation, why should we assume that Hobbit DNA tests presenting only H. sapiens DNA were contaminated? No, the debates will continue, and, as anyone who both understands and respects the scientific process recognizes, they should not be discouraged.

Regardless, each of these esteemed professionals has already contributed a great deal to the resolution of some of the most exciting and momentous issues in anthropological history. Who were the Hobbits? From where and what did they come? Has nature selected for human intelligence by more than one means? At least some of the new species proponents have lately deemphasized their claims of descent from H. erectus and of endemic dwarfing in favor of an earlier, small-bodied and small-brained ancestor that perhaps departed from Africa long before H. erectus. For the first time, many believe that a more primitive but nonetheless highly intelligent species of human recently coexisted on earth with H. sapiens for tens of thousands of years.

The known LB specimens have been analyzed exhaustively, at least by those who continue to control their remains. DNA tests have been attempted, but thus far they have proven inconclusive. So we need to keep digging for clues on Flores and elsewhere. And that’s exactly what Morwood intends to do. He’ll return to the Sao Basin shortly and, of course, to the LB cave. In Timor’s Atumba Basin, he has already uncovered evidence of 120,000-year-old hominins who “could be early modern humans,” he says, or, on the other hand, “late representatives of the Hobbit lineage” (Morwood 2009). He hopes to excavate various Southeast Asian archipelagoes as well, the Sulawesi island’s Maros region and Walanae Basin in particular.

Much work remains to be done, and hopefully most of the name-calling, side-taking, and wagon-circling will be left far behind us in favor of camaraderie, fair competition, and an irrepressible passion for the truth. If so, the best Hobbit news is yet to come, and I expect that at least some of our queries will be resolved in the very near future. But whatever the final verdict, as Jungers recently said to me, “The answer will be profound no matter what it is.” l

©2009 Kenneth W. Krause, all rights reserved


Kenneth W. Krause

Kenneth W. Krause is a contributing editor, books editor, and "The Good Book" columnist for the Humanist and a contributing editor and columnist for the Skeptical Inquirer.