Geist 1978; Merker 1984
Classification: Feeding Models
Mnemonic: Hand to Mouth
Specific Model: Stalking behind trees.
Original Proponent(s): Geist (1978) Merker (1984)    
Assessment: Popularity: Feeding Models were ranked 2nd (out of 9) most popular in the texts reviewed. No text referred to this model explicitly.
Simple: #35  / 42 (47%)
Detailed: #21 (=2)  / 42 (55%)
Basic Summary: Stalking (quietly approaching prey) in an arboreal setting was a major factor in the evolution of hominin bipedalism.    
Discussion: The fact that human bipedalism is relatively slow and vulnerable compared to most terrestrial animals has not prevented hunting models from being constructed which are based on other characteristics of human bipedalism which can be construed in a more favourable light. The stalking model is one such idea. It’s premise is the observation that the human foot is very well adapted to stalking behaviours, as it provides very sensitive tactile feedback, allowing a would-be hominin hunter to creep up within grabbing distance of prey without being noticed, before stunning it with hand-held rocks.

The idea’s strengths are that it supports and is supported by suggestions of a more BHBK-like gait amongst early hominids and the associated greater arboreality that goes with it. It also draws support from the fact that a vertical posture gives better camouflage in woodland.

The model is supportive of carrying models in that it suggests that bringing rocks to the prey would also make a killing more plausible. Valerius Geist, a Canadian Biologist and specialist in mountain sheep (which do not have such tactile feet as we do), developed the two-phase model in his book “Life Strategies, Human Evolution, Environmental Design” (Geist 1978).

Along similar lines to Elizabeth Vrba (1985) and Jonathon Kingdon (2004), he begins with the observation that many African mammalian lineages seem to be characterised by more ‘primitive’ forest forms and more ‘advanced’ open habitat forms. (It should be noted here that by ‘advanced’ he is referring to forms with more derived traits and fewer conservative ones, changes in sexual dimorphism and a general increase in body size.) He states his belief, from the onset, that “our genus evolved much as did other lineages of large Pleistocene mammals” (Geist 1978:212) and that, consequently, the genus Pan appears “so close to the origin of the hominids that it is a form frozen in its evolutionary development since the middle Tertiary...” Geist (1978:213.) This is quite a commonly held position and Geist adds to it another much-stated view: “The next step away from the forest-adapted Pan-like ancestor is expected to be in the direction of grasslands…” but, interestingly, he qualifies it with a caveat, “…beginning with the ecotone between the forest and steppe, namely the savannah and riparian vegetation communities” and specifically suggesting that “the availability of river valleys and highly productive floodplains with marsh and plains vegetation studded with well-spaced trees or clumps of shrubbery” in gallery forests should not be forgotten (Geist 1978:213-214.)

After a brief review of the current debate about australopithecine dimorphism, he generally comes down to the view that there were two, closely related, contemporaneous forms, the robust and the gracile, to set the paleoecological context of his model, which he postulates was cantered in and around a habitat which he labelled “the wet savannah” (Geist 1978:215.)

Geist then goes on to consider exactly what kind of adaptive specialisations such a hominid might be expected to have in such a habitat, always alluding to how this might be adaptive to the evolution of hominid bipedalism.

He writes: “What are the characteristics of moist savannah and floodplain and riparian plant communities? Marsh communities show a great productivity gradient (Verner and Willson 1966); they have a great diversity in plant and animal life, and they are less subject to oscillations in the availability of food than are terrestrial communities, due to the continuous presence of water. Furthermore, the photosynthetic layer is greatly compressed compared to the typical forest ecosystem, and consequently the forage resource is compressed and concentrated. This makes the production of the photosynthetic layer almost totally available to an ape-sized primate, and permits him to reach higher population densities than in the forest ecosystem whose photosynthetic layer it can only partially exploit. Let us assume an arboreal omnivorous primate begins to adapt to the marsh and savannah communities along water courses. He is faced by the following: a seasonally fluctuating foraging area, the size of which varies with the flooding of rivers and the annual rain cycle…
Geist (1978:215-216.)

Interestingly, Geist rejected the energy efficiency model (probably only because his paper pre-dated the one by Rodman & McHenry (1980) showing that, at slow speeds, bipedalism is actually slightly more efficient than quadrupedalism) even though he followed Robinson (1973) in postulating that the gracile forms lived in more open (and hence less food rich) habitats than did the robust forms. Most of his arguments for the first phase, or origin, of hominid bipedalism are complimentary, if not identical to, to the postural feeding model of Hunt. A key aspect being that the gallery forest, or mesic savannah, habitat has a photosynthetic layer that is compressed such that a large hominid could reach practically all of it simply by standing up. He also pre-empted Wrangham (1980) in suggesting that it would have been more efficient to remain upright whilst shuffling from bush to bush gathering foods than to repeatedly switch back and forth between quadrupedal locomotion and postural bipedalism.

Geist also supported vigilance ideas:
“It would, of course, also permit predators to be spotted, a common claim for the evolution of bipedality (Robinson 1972a, Kortland 1967) However predatory spotting would not require continuous bipedality; it would be required for foraging among relatively tall, fragile herb stalks and within the thin, often thorny branches of low shrubs.”
Geist (1978:216)

So for the first phase – the initial incentive for australopithecines (assumed by Geist to be ancestors of Homo) to begin moving bipedally - Geist is indistinguishable from Hunt and others who have proposed that postural feeding in forest edge habitats were the main factor to cause hominin divergence.

Later, however, in the next chapter “The Homo erectus Stage”, Geist supplements this with a specific and quite unique hunting phase which proposes slow stalking was they key behaviour being selected for the evolution of modern human anatomy. The basis for his idea is observation that the human foot, especially compared with the ungulates he knew best, was very tactile and sensitive allowing humans, he postulated, to creep very quietly near to prey without detection.

Geist put it this way:
“Far from having been reduced to the function of locomotion, our feet have evolved into instruments of silent approach to permit the capture of a prey unaware of the hunter”
Geist (1978:252.)

“Here appear to lie the evolutionary origins of our most unusual legs and feet, not only in the economic deployment of legs during walking. The human leg and foot appear to be the product of systematic hunting… Although other carnivores, such as the large cats (Schaller 1972, Hornocker 1970), are also stalkers, they possess a technique not practiced by humans. This is a final rush over some distance that rapidly closes the gap between predator and prey. The human form of stalking brings the stalker literally within arm's reach of its prey. It is for this reason, because we lack the lightening-fast rush at the end of the stalk, that sophisticated silent stalking is at a premium.”
Geist (1978:253).

The first critical thought one might have to such a suggestion is to question exactly how such early hunters, no matter how quiet and well adapted they were to stalking, would manage to get within “arms’ reach of its prey” without being spotted. One answer suggested by Merker (1984), is that their upright posture, being partially hidden behind trees, would have allowed them to get much closer than a typical quadrupedal hunter. But Geist argued that, compared to the robust form of australopithecine, it was the gracile form (our putative ancestors) that lived in more open habitats and that during this phase of evolution, they were not hunters. The hunting phase Giest is proposing is firmly with much later Homo forms in even more open habitats. Putting aside arguments ad absurdum the model does offer an explanation for human sensitivity in their feet, a feature that must be quite unusual in the animal world. As Geist’s model is really a two phase model, and the first phase is similar to Hunt’s (1994), only the later part is evaluated here.

A short paper to the American Anthropologist by Bjorn Merker (1984) supported this stalking model for the evolution of bipedalism and, importantly, offered to help answer one of it’s biggest question marks against it: Namely how is it proposed that prey would not have seen stalk-hunting hominids approaching? His suggestion is that upright posture is ideally suited for this in the context of wooded habitats. Although this somewhat contradicts Geist’s writings – he is quite clear that such hunting phases happened later in human evolution, away from wooded habitats – it may help, if one suggests that stalk-hunting might have been practiced around wooded water courses, such as gallery forests.
 Strengths: Compatible with paleocological evidence of showing that 'wet savannah' (i.e. ripparian woodland) was a habitat conducive the the evolution of early humans. Attempts to explain the peculiar tactile sensitivity of the foot.    
Weaknesses: Lacks plausibility. Getting within arm's reach of some prey is difficult enough for highly camoflaged and highly adapted carnivores let along a unpright hominin that is not camoflaged at all.    
1.1 Survival Value 6 (Fair) Assuming that hominids could, indeed, stalk prey as well as Geist seems to be suggesting – getting to within arm’s length with a rock in hand before detection – would, clearly give very strong selection for bipedal locomotion. However, the plausibility of the basic premise seems to be a little suspect: ungulates would surely flee long before any would-be attacker got anywhere near arms’ length.    
1.2 Sexual Selection 5 (Fair)    
1.3 Not Teleological 6 (Fair) One of the strengths of Geist’s model is that it is multi-phased. Each phase offers an evolutionary advantage through an incremental step towards human bipedalism that would be beneficial at that stage    
2.1 Improved Food Acquisition 7 (Good) Although the model is based upon improved food acquisition. It was judged not as highly as other models in this category because stalking prey seemed less plausible than the others.    
2.2 Accounts for Predation 3 (Poor) The model is postulating that human ancestors were living in open habitats competing for prey with very large, well adapted carnivores. It is not made clear how it is proposed hominids remained the hunting and avoided becoming the hunted.    
2.3 Why Apes are not Bipedal 6 (Fair) If we assume Geist first phase is correct that early hominids lived in gallery forests and/or mesic savannah habitats where postural feeding would have given them a distinct advantage, then his model would explain the ape-human divergence quite well. However the problem really lies with the later phase of the model.    
2.4 Extant Analogues 2 (Poor) Although postural feeding has been observed in extant apes (Hunt 1994), stalk-hunting has not.    
2.5 Applies to Both Sexes 6 (Fair) Geist seems to assume that females would not participate in stalk-hunting, although there would appear to be no reason why they shouldn’t considering the most important characteristic being suggested here is quietness on foot.    
3.1 Hominid Anomalies 2 (Poor) Geist does not attempt to explain post-cranial anatomical anomalies of the australopithecines    
3.2 Fits Paleoecological Record 6 (Fair) The model does fit the accepted paleoecological record since the Miocene, reasonably well, although it is appears to be too firmly seated in the savannah paradigm, namely that human evolution was characterised as a progressive shift from forest to savannah, to the steppe    
3.3 Precursor to Strider and knuckle Walker 6 (Fair) Stalking, requiring both upright posture and slow upright gait does overlap considerably with human bipedalism.    
4.1 Extended Explanatory Power 4 (Fair) The stalk hunting idea is only a part of Geist’s speculations on human evolution whose scope is very broad indeed. Specifically here, the suggestion is that stalk-hunting would explain the evolution of human feet and legs    
4.2 Complimentary 6 (Fair) Complementary to most arboreal, carrying or weapon wielding models. Incompatible with those relying on open habitats e.. for vigilence.    
4.3 Falsifiable or Testable 4 (Fair) Geist’s speculations are usually complimented with predictions and tests, although the major one being evaluated here, the stalking model, is rather lacking in that area.    
References Geist, V (1978). Life strategies, human evolution, environmental design. (Towards a Biological Theory of Health). Springer-Verlag (New York)
Kingdon, J. (2003). Lowly Origins. Princteton University Press (Woodstock).
Hunt, K. (1994) The Evolution of human bipedality: ecology and functional morphology. Journal of Human Evolution 26:183-202.
Merker, B (1984). A Note on Hunting and Hominid Origins. American Anthopologist Vol:86(1) Pages:112-114
Vrba, E (1985). Ecological and Adaptive Changes Associated with Early Hominid Evolution. In: Delson, Eric (eds.), (1985). Ancestors: The Hard Evidence. Alan Liss Inc (New York)