The honeyguide and the people who follow it

A wild bird finds bees' nests. People cannot open them alone, and the bird cannot open them at all. Across parts of Africa the two call to each other, travel together and split the result — and unusually for a story this good, it has been tested experimentally.

BeesReviewed 2026-08-20

What a honeyguide is

The greater honeyguide, Indicator indicator, is a drab starling-sized bird found across much of sub-Saharan Africa. Two things about it are unusual. It eats beeswax, which almost no other vertebrate can digest. And it is an obligate brood parasite: it lays in the nests of other hole-nesting birds, and the newly hatched chick, equipped with sharp hooks on its bill, kills the host's young.

Neither of those facts is the reason people have heard of it. It is known because it leads humans to bees' nests, and because it appears to do so deliberately.

What guiding looks like

  1. The approachThe bird flies close to a person, gives a distinctive chattering call and moves conspicuously, fanning its tail. It is unmistakable to anybody who knows it and easy to miss for anybody who does not.
  2. The leadIt flies a short distance in the direction of a nest, perches in view, and waits. When the person follows, it moves on. The sequence repeats, sometimes over a kilometre or more of broken country.
  3. The arrivalThe bird falls quiet and perches near the nest tree. The perch heights and the calling pattern change as the destination approaches, which is what allows an experienced hunter to tell that they have arrived.
  4. The openingThe people smoke the colony, cut into the cavity and take the honeycomb. The bird waits, then feeds on the wax, the brood and the fragments left behind.

Neither party could do this alone. The bird can find nests and cannot open them: a wild colony in a tree cavity is beyond a small bird with no way through the wood and no defence against thousands of stings. The people can open nests and are much worse at finding them. The measured effect of following a bird is a substantial reduction in search time and a large increase in the rate of successful finds.

The calls, and why they are the interesting part

Honey hunters do not simply follow a bird that turns up. They call it. In the Niassa region of northern Mozambique, Yao hunters use a trill followed by a grunt, learned from their fathers and used for nothing else. Hadza hunters in Tanzania use a whistle. Boran gatherers in northern Kenya use a different signal again.

The obvious question is whether the birds are responding to the specific sound or simply to human presence. It has been tested. A playback experiment in Mozambique compared the Yao trill against control sounds matched for acoustic structure — including human words spoken in the same voice and the calls of other animals. The traditional call roughly doubled the probability of being guided and more than doubled the probability of being led to a nest. The birds were responding to the signal, not to the person.

A later experiment took the design across regions and found that birds respond most strongly to the signal used by their local human community, and less strongly to the equivalent signal from elsewhere. That is evidence that the bird learns a local convention rather than an innate human-shaped sound, which puts a learned communication tradition on both sides of the partnership.

What the bird gets, and the argument about it

The straightforward account is reciprocity: the bird leads, the people open the nest, the bird eats the wax. Wax is a real reward — honeyguides digest it, and few animals compete for it — and in several communities hunters deliberately leave comb behind for the bird.

It is not universal. Hadza hunters have been recorded deliberately withholding wax, burning or burying the remains, on the explicit reasoning that a hungry bird guides again tomorrow. That is the opposite of reward and the mutualism survives it, which tells you the relationship is more robust and less tidy than the storybook version. Both behaviours are documented, in different communities, by researchers who went looking.

Who works with honeyguides

  • Yao honey hunters in northern Mozambique, where the reciprocal-signalling experiment was carried out.
  • Hadza foragers in northern Tanzania, among whom honey is a large share of the diet and guided finds are a measurable part of it.
  • Boran gatherers in northern Kenya, the subject of the first quantified field study of guiding in the 1980s.
  • Awer, Ogiek and several other communities across eastern and southern Africa, in accounts of varying age.

A partnership going quiet

Guiding has lapsed across much of its former range. Where honey hunting declines — through forest loss, protected-area exclusion, the availability of sugar, and young people leaving for other work — the calling stops, and the birds appear to stop guiding. The behaviour is learned, and a learned behaviour with no one left to learn it from disappears.

That makes this a conservation problem of an unusual shape. What is at risk is not a species — the greater honeyguide is not endangered — but an interaction, held in the memory of two populations at once, and lost as soon as either side stops.

The honey badger claim

The two claims are constantly reported in one breath because they sound alike, and the effect is to make a rigorously tested finding look like folklore. Keeping them apart is the single most useful thing a page on this subject can do.

The claims on this page

Every substantive claim is placed on a tier, and the top two tiers must also state what they are not claiming.

Emerging research

Real studies exist and point somewhere, but they are small, in vitro, in animals, or not yet replicated. Direction is not the same as a result.

Emerging research

Greater honeyguides respond to specific human recruitment calls, and respond more strongly to the signal used by the local human community.

Two playback experiments with matched acoustic controls, in Mozambique and then across regions, show the traditional call substantially raises the probability of being guided, and that birds discriminate between local and non-local human signals.

What this does not claim: It does not establish that the bird understands the call as a request, that it intends to cooperate, or that any equivalent partnership exists with a non-human species. It establishes a learned association strong enough to survive controlled comparison.

  • Reciprocal signaling in honeyguide-human mutualismScience (2016) · Peer-reviewed article
  • Culturally determined interspecies communication between humans and honeyguidesScience (2023) · Peer-reviewed article

Commonly claimed, but not supported

Widely repeated and not supported by current evidence. Listed so the correction is here rather than only the claim.

Commonly claimed, but not supported

Honeyguides lead honey badgers to bees' nests.

Long repeated in reference works and broadcast media, and supported by very few verified direct observations. Reviews of the evidence do not find it established. Reported here because a reader will meet the claim, not because this site holds it.

Sources

  • Reciprocal signaling in honeyguide-human mutualismScience (2016) · Peer-reviewed article
  • Culturally determined interspecies communication between humans and honeyguidesScience (2023) · Peer-reviewed article
  • Honeyguides and honey gatherers: interspecific communication in a symbiotic relationshipScience (1989) · Peer-reviewed article
  • Mutualism and manipulation in Hadza-honeyguide interactionsEvolution and Human Behavior (2014) · Peer-reviewed article
  • The World History of Beekeeping and Honey HuntingRoutledge (1999) · Historical text
  • HoneyHQ editorial synthesisHEKNO Ltd · HoneyHQ synthesis

How sources are selected and weighted is set out in the sources and evidence policy.

Documented traditions

Specific practices behind this subject, each naming where it is documented, among whom, and whether anyone still does it.

Read next

Tools that use this

HoneyHQ saves your cookie choice on this device. Optional analytics run only if you accept — set a preference now, or change it later on the cookies page.