The Origin of Science: The Evolutionary Roots of Scientific Reasoning and its Implications for Citizen Science

1. What they did

Liebenberg builds on his 1990 book, The Art of Tracking: The Origin of Science, to argue that the art of tracking may be the origin of science. He sets out to resolve an old puzzle in human evolution: how did the human mind evolve the ability to reason scientifically, if hunter-gatherer life did not seem to need it?

The book has twelve chapters. They cover his near-fatal 1990 kudu chase with !Nate, Kayate and Boroh//xao at Lone Tree; hunter-gatherer subsistence; persistence hunting; the art of tracking as practised by Kalahari trackers; how tracking may have evolved; an evolutionary model of science; the scientific imagination, comparing tracking with Einstein’s physics; science, language and art; modern tracking and CyberTracker; citizen science; and the future.

It draws on his fieldwork with !Xõ and /Gwi trackers in Botswana from 1985 onwards, interviews with former hunters in Namibia’s Nyae Nyae Conservancy in 2009, his tracker evaluations, the CyberTracker project, and the scientific literature.

2. What they found

Persistence hunting

• Hunts take place in the hottest part of the day (about 39 to 42 °C). The hunters track the animal at a running pace and keep chasing it before it can cool down in the shade, until it is run down.
• Of eight attempts recorded (1985, 1990, and with film crews in 1998 and 2001), three kudu were killed. Leaving out hunts attempted only for filming, four of five attempts succeeded (80%).
• In 1998 a hunt covered about 35 km in 3 hours 35 minutes (by vehicle odometer). Two hunts in 2001, recorded with GPS, covered 25.1 km in 3 hours 50 minutes (6.3 km/h) and 33 km in 4 hours 57 minutes.
• Losing the trail was the main reason hunts failed. The hardest moments were when the animal rejoined its herd, circled back over its own tracks, or other antelope crossed the trail.
• Lone Tree hunters said steenbok, duiker and gemsbok can be run down in the rainy season, and kudu, eland and red hartebeest in the dry season, when loose sand tires them. Kudu bulls tire faster than cows, and a tiring kudu often breaks away from its herd to hide.
• The hunter must keep comparing his own body with the kudu’s, judging its tiredness from its stride and kicked-up sand. Liebenberg failed to do this and nearly died of heat stroke; !Nate risked his own life running back to camp for water to save him.
• Over 46 days of hunting without dogs (1985 to 1992), 41 bow-and-arrow attempts gave only two kills. Persistence hunting gave a higher meat yield than the bow and arrow, about the same as snaring. Only hunting with dogs gave a markedly higher yield.
• Karoha Langwane may be one of the last persistence hunters.

The art of tracking

• Kalahari trackers can identify individual people, and individual large animals such as kudu, gemsbok and eland, from their footprints. !Xõ trackers identify the tracks of most animals larger than mongooses, and picked out the Namaqua dove from Liebenberg’s drawings of dove tracks. Women were as knowledgeable as men.
• Mistakes come from looking too quickly. A tracker who called a small bird’s track a “mouse” said afterwards that you must study tracks carefully and think before deciding.
• To recognize signs, a tracker must know what undisturbed ground looks like. A preconceived image of a sign helps, but can make you “see” what you expect. Difficulty depends on how much information a sign holds, how sparse the signs are, and how many other animals’ signs are nearby. Knowing when there are no signs where there should be is also important.
• Trackers describe a “burning sensation” when their quarry is near, and hair rising when danger is close. Liebenberg links this to peripheral perception. While he was evaluating Joseph Mabunda for his Senior Tracker certificate in Thornybush, Joseph sensed hidden lions moments before he saw them, probably from their flicking ears.
• Ageing tracks and signs is among the hardest skills. Estimates are better for fresh signs, but approximate estimates are good enough, because trackers keep re-checking as they close in.
• Trackers build a working reconstruction of what the animal was doing, partly from the spoor and partly from their knowledge of behaviour, terrain and time of day, and revise it as new signs appear. Examples: !Nam!kabe found a resting wildebeest near its feeding ground; ranger Karel Kleinman read a male lion’s tracks as going to find a female, and confirmed it.
• Systematic and speculative tracking: systematic tracking gathers signs one after another and rests on experience (inductive-deductive reasoning). Speculative tracking forms a working hypothesis and looks for signs where they are expected (hypothetico-deductive reasoning). Most learners start as systematic trackers; the move to speculative tracking needs a lot of experience and can be very difficult, and the longer someone does only systematic tracking, the harder it becomes. Systematic tracking suits open sandy ground, small animals that circle within a small home range, and calves. Speculative tracking suits hard or rocky ground, thick woodland, and large animals travelling far. Even experts may find speculative tracking hard in unfamiliar areas. With dangerous animals or criminals, you look ahead first for danger, then predict the route, and only then look for tracks.
• Trackers’ knowledge of animal behaviour is anthropomorphic but works well in practice. For example, they read jackal calls to tell whether a jackal is following a hyena or a lion.
• /Gwi trackers have eleven names for ants, termites and velvet ants. Much of this knowledge goes far beyond hunting needs: knowledge for the sake of knowledge.
• Citing Richard Lee’s data on Ju/’hoansi hunters aged 15 to 38, 70% of kudu kills were made by 17% of hunters, and almost half made none. Liebenberg links this to the intellectual demands of tracking rather than eyesight or fitness.
• Trackers explained hoof shapes by how antelope escape danger: pointed hoofs for fast runners in the open (steenbok, springbok, gemsbok), rounded hoofs for woodland species that dodge through bushes (duiker, kudu, eland). One tracker gave this explanation in 1985 and four others gave it independently in 2012.
• Tracking works like a scientific research programme. Sharing meat gave hunters the freedom to be wrong at times. In evaluations, even experienced trackers sometimes misidentified a track at a glance, but after discussing it together they reached the correct answer, even when only one of them had been right at first.
• Belief in myths varies between trackers. !Nate rejected an old sun myth for lack of evidence and invented his own theory of how the sun moves.

The evolution of tracking

• Liebenberg distinguishes simple tracking (following clear footprints in easy conditions), systematic tracking, and speculative tracking.
• He suggests tracking began where it is easiest, such as the open southern Kalahari dunes, and grew more skilled as conditions became harder: denser grass and woodland, or harder ground. Climate swings between dry and wet periods may have selected for better tracking.
• Systematic tracking may have evolved with Homo erectus about two million years ago; speculative tracking, and with it creative science, by at least about 70,000 and possibly more than 100,000 years ago.
• The non-arbitrary link between a seen track and the imagined animal suggests that human visual imagination may have begun with reading tracks and signs.
• Bell-shaped figures in El Castillo cave, Spain, may show hoofprints of an animal jumping in soft mud.

Science and the scientific imagination

• Liebenberg proposes an evolutionary definition of science: scientific reasoning as an adaptation that evolved through natural selection. Knowledge has grown along a logistic curve over more than 100,000 years, rather than through sudden revolutions.
• Finding plant foods can rest on experience alone; tracking moving animals needs creative hypotheses.
• From caracal tracks at Cape Point, Liebenberg worked out that the caracal had leapt straight up to catch a bird in mid-air, something he did not know caracals could do. The literature later confirmed that they can.
• He compares a tracker’s projection into the animal with Einstein imagining pursuing a beam of light.
• Superstition may be a by-product of the creative imagination: “hypotheses” that cannot be tested.

Modern tracking

• Stander and colleagues (1997) found a Ju/’hoan team correct in 98% of 569 reconstructions from spoor.
• Evans and colleagues (2009) found experienced otter surveyors misidentified 37% of otter tracks and called 26% of “otter-like” tracks otter. Average scores rose from 61% to 79% at a second evaluation three months later.
• CyberTracker certification recognizes seven levels, from Tracker I to Master Tracker. Each candidate is evaluated in the field, asked for their own interpretation first, and has mistakes corrected and explained throughout.
• Mark Elbroch earned his Senior Tracker certificate in South Africa and then started CyberTracker evaluations in the USA.
• The CyberTracker software began in 1996 and lets trackers who cannot read or write record data with icons. Karel Benadie and James Minye recorded up to 266 observations in a day, identified individual black rhinos by the crack patterns in their tracks, and showed seasonal changes in rhino feeding, published in Pachyderm in 1999. Kruger National Park used 210 units and gathered over 1.97 million records from 2004 to 2006.
• In 2008, !Nate and Karoha made careless mistakes because they no longer hunted often. After two years of CyberTracker surveys, all four trackers evaluated in 2010 excelled, and they received the Traditional Master Tracker certificate that year.

Citizen science and the future

• Citizen science ranges from participatory projects run by scientists to independent citizen scientists publishing their own work. Liebenberg describes his own path as an independent researcher.
• He suggests building credibility first through reliable data.
• He calls for free access to scientific literature, and argues that science needs ethics and compassion.
• Two Kalahari elders described a conservation ethic: during droughts, hunter-gatherers left scarce plants to recover and hunted more, then stopped hunting after good rains so animal numbers could recover.

3. What they concluded

• The art of tracking may be the origin of science. Speculative tracking uses the same hypothetico-deductive reasoning and intellectual abilities as modern science, so science may have evolved more than 100,000 years ago.
• The differences between tracking and modern science are mainly technological and sociological. Modern science has grown mainly in quantity, not quality.
• Scientific reasoning may be innate, so anyone, whether or not they can read or write, can contribute to science.
• Traditional trackers can be employed in research. Because they no longer depend on tracking to survive, objective evaluations are needed to tell good trackers from poor ones and so validate the data they collect.
• Science needs to be freely accessible, and tempered by ethics and compassion.

4. How tracking was or was not used

Tracking is the subject of the book. The author tracked with Kalahari hunters from 1985 onwards. He describes persistence hunts, bow-and-arrow hunts and lion tracking, and recorded routes of persistence hunts with GPS in 2001. He gives examples from his own tracking: an aardvark over rocky mountains in the Klein Karoo, and caracal tracks at Cape Point. He interviewed trackers and former hunters, and draws on his tracker evaluations and the CyberTracker project.

5. Tracker or observer certification

• The book describes the CyberTracker Tracker Evaluation system: seven levels from Tracker I to Master Tracker, assessed by an accredited Evaluator in the field.
• Certificates mentioned: Joseph Mabunda’s Senior Tracker evaluation in Thornybush; Mark Elbroch’s Senior Tracker certificate; and in 2010, !Nate, Karoha, Njoxlau and /Uase were awarded the Traditional Master Tracker certificate.
• The trackers in the author’s earlier fieldwork were traditional hunter-trackers, before formal evaluation existed. Liebenberg notes that hunting ability varies widely between individuals.

6. Other technologies, skills or innovations

• CyberTracker icon-based data collection for people who cannot read or write: first on an Apple Newton handheld wired to a Garmin GPS, later on smartphones. Data are shared through networks such as DataONE.
• GPS used to plot persistence hunt routes.
• Suggests combining tracking with radio tracking and camera traps.
• Traditional skills described: bows and poison arrows, snares and traps, the springhare probe, smoking animals out of burrows, and hunting with dogs and horses.
• Hunting with dogs does not replace tracking skill. Dogs reached the Kalahari in the late 1960s and changed hunting dramatically. Even with dogs, hunters must understand tracking, animal behaviour and the environment. They read spoor from previous days to direct the dogs, because dogs only react to scent and sounds close by; on their own, dogs only run down small animals such as bat-eared foxes flushed in front of them, and chase anything that moves. Hunting with dogs, and from horseback, is much more efficient, but does not need the skill, expertise and ingenuity of persistence or bow-and-arrow hunting. Liebenberg links the decline in Kalahari tracking skills to dogs and horses replacing the bow and arrow and persistence hunting, “since the dogs are used to do the tracking.”

7. Implications for conservation (from the paper)

• Trackers can contribute to research, monitoring, ecotourism, anti-poaching and crime prevention, and can extend researchers’ capacity to gather data by orders of magnitude. Employing them brings income to local communities and helps keep tracking skills from being lost.
• Knowing where rhinos drink and sleep helps anti-poaching patrols focus where poaching is most likely.
• CyberTracker data in Kruger National Park support patrol planning, early warning of disease, tracking poacher entry points, invasive species control and fire management.
• Liebenberg hopes for a worldwide environmental monitoring network, with communities monitoring their own environments.