Fork-tailed Drongo — Deceptive Alarm Mimicry

The fork-tailed drongo (*Dicrurus adsimilis*) is the paradigm case of deceptive referential signaling in a non-mammalian species. It mimics the alarm calls of other species — including meerkat, southern pied babbler, and birds of prey — to cause those species to flee, then steals their food. It issues genuine alarm calls as well, maintaining just enough honest signaling to preserve its credibility for the next deception.

What the Signal Is

The drongo produces both genuine and false alarm calls. The false alarms are mimicry: the drongo reproduces the alarm calls of the target species, or of the predators that species fears. Documented repertoire includes:

• Alarm calls of meerkats
• Alarm calls of southern pied babblers and sociable weavers
• Calls of birds of prey and owls
• Calls of bocage's bushshrike, thrushes, tchagras, bulbuls
• Cat mewings

The drongo spends approximately 29% of its time trailing other foraging animals — particularly ground-foragers who disturb insects and are thus both productive to follow and easy to displace.

The Deception Mechanism

When a target finds food, the drongo issues a false alarm. The target flees. The drongo drops down and takes the food. Key features of the system:

Post-deception repair: After a false alarm, the drongo issues an "all clear" signal to minimize disruption to the target's foraging and to rebuild trust — allowing the deception to be repeated. This is sophisticated: the drongo is managing its reputation within a long-term relationship.

Flexibility: The drongo varies between genuine alarms, false alarms, and direct attacks. Studies suggest it selects based on context — switching to direct attack after a failed false alarm attempt. Tom Flower (2014, *Science*) showed it even varies *which species' call* it mimics within a single interaction, preventing habituation by receivers.

Target selection: Primary targets are ground-foraging species — meerkats, southern pied babblers, sociable weavers, wattled starlings. The drongo lacks the morphological adaptations to efficiently hunt the larger, more calorie-dense prey these species find; kleptoparasitism is a bill-morphology workaround.

Scale of the Deception

Approximately 25% of drongo food intake comes from kleptoparasitism; an additional 10% from prey flushed by associated species. Prey acquired through kleptoparasitism is typically *larger* than self-foraged prey. However, kleptoparasitism appears concentrated during food scarcity (drought, cold days) — suggesting it is a fallback, not a primary strategy.

Honest or Deceptive — The Full Picture

Marked as honest: false because the signal's primary function in this context is deception. But the drongo's signaling is genuinely mixed:

• Genuine alarms: the drongo issues real alarm calls for real predators, which benefits its foraging associates (and the drongo itself, as sentinel behavior attracts followers)
• False alarms: the deceptive subset, used opportunistically
• The ratio is not known but must be high enough to maintain receiver responsiveness — this is the core evolutionary constraint on deceptive signaling. If false alarms become too frequent, receivers stop responding and the strategy collapses.

This is a direct empirical refutation of the "honest signaling only" equilibrium predicted by simple signaling theory. Stable mixed strategies with some deception are possible when the deceiver is rare enough not to erode receiver trust.

Theory of Mind Question

Researchers have raised the possibility that the drongo's flexible deception — varying calls, managing post-deception repair, adjusting tactics based on receiver response — implies something approaching a theory of mind. This remains undemonstrated and disputed. The behavior could be explained by stimulus-response learning without any representation of the receiver's mental state.

Key Researcher / Discovery Story

Tom Flower (University of Cambridge, then UCT) documented the false alarm mimicry system in detail through Kalahari fieldwork, publishing the core finding in *Proceedings of the Royal Society B* (2010) and extending it in *Science* (2014) to show flexible dialect switching. The Science paper was notable for demonstrating that the drongo actively avoids repeating the same false alarm to the same individual — varying its mimicry to prevent habituation.

Vault Cross-Links

The drongo's targets include animals that may use vervet-like referential alarms — this is a system where the deception exploits the fact that alarm calls ARE referential. Deception requires a referential system to exploit. See vervet-monkey-alarm-calls, diana-monkey-alarm-calls. The drongo is itself parasitized by common-cuckoo-brood-parasitism (African cuckoos use 21.8% of drongo nests).