Prairie Dog Alarm Calls — Slobodchikoff's Vocabulary Hypothesis
Con Slobodchikoff (Northern Arizona University) argues that black-tailed prairie dog (*Cynomys ludovicianus*) alarm calls encode specific predator descriptions — not just predator class but size, color, speed of approach, and shape. If the strongest version of his claims holds, prairie dogs possess the most descriptively rich referential communication system yet documented outside of humans or great apes.
What the Signal Is
A high-pitched vocalization produced upon predator detection. Acoustically it appears as a single call type, but Slobodchikoff's spectral analysis reveals variation in frequency modulation, harmonic structure, and call duration that he interprets as encoding distinct predator properties.
The jump-yip ("jump yip") display is a distinct signal: a prairie dog stretches vertically and throws its forefeet skyward while calling. A single jump-yip triggers a wave through the colony — neighbors jump-yip in sequence. Slobodchikoff interprets this as an "all-clear" or vigilance check; the wave's propagation distance signals how many neighbors are actively watchful, which the instigator uses to calibrate its own foraging risk.
What It Encodes — Slobodchikoff's Claims
According to Slobodchikoff's research:
- Predator identity — different calls for coyote, human, hawk, dog
- Size — call structure correlates with body size of the predator
- Color — different call elements for differently colored human clothing (tested by walking humans in different-colored shirts past colonies)
- Speed of approach — faster approach → different call structure
- Shape — circular vs. triangular objects approaching elicited different calls
Slobodchikoff argues these properties are compositionally combined — effectively a form of grammar. He further claims prairie dogs have calls for *non-predators*, suggesting the system is generative rather than fixed to a predator library.
Controversy
Slobodchikoff's claims are at the strong end of the referential spectrum and have not been independently replicated to the same degree as the vervet-monkey-alarm-calls findings.
Key objections:
- The color/shape experiments rely on fine acoustic distinctions that may reflect arousal variation rather than discrete encoding
- The "grammar" claim requires demonstrating that elements can recombine to produce novel meanings — this has not been shown with the rigor applied to Campbell's monkey (see campbells-monkey-suffixation when written)
- Playback studies confirming that receivers extract the specific predator information (analogous to Seyfarth & Cheney's vervet design) have been harder to execute for prairie dogs
The weight of evidence supports that prairie dog calls are more informationally rich than simple arousal-level signals — but whether they reach "vocabulary" remains disputed.
Jump-Yip: A Different Signal
The jump-yip is among the better-characterized prairie dog signals. It propagates as a social wave, and the wave length is functional: a longer wave means more vigilant neighbors, which frees the instigator to forage more. This is honest signaling via a distributed colony-level computation. It is not referential (it doesn't encode external predator properties) but it is cognitively interesting — the instigator is reading the colony's collective vigilance state.
Honest or Deceptive
Marked as mixed because:
- Predator alarm calls appear to be honest for the standard reason (false alarms degrade receiver responsiveness, which harms the caller)
- Whether alarm calling is altruistic or selfish is debated — calls may increase the caller's own predation risk (makes the caller conspicuous) or decrease it (predator targets panicking conspecifics instead)
- Evidence supports kin-selection as the primary driver: prairie dogs call more when kin are nearby
Key Researcher / Discovery Story
Constantine (Con) Slobodchikoff began this work in the 1980s–1990s at Northern Arizona University and published extensively through the early 2000s. His 2012 popular book *Chasing Doctor Dolittle* made the color-encoding claims widely known. The work is genuine and methodologically careful but has attracted sustained skepticism from linguists and animal communication specialists who argue the grammar inference goes further than the acoustic data supports.
Open Questions
- Can independent replication confirm the color and shape encoding?
- Is there any cross-population variation in call structure that would suggest cultural transmission of predator labels?
- What is the receiver's mechanism — do prairie dogs respond differentially to played-back calls for "large fast predator" vs "small slow predator"?
- The jump-yip wave as distributed vigilance sensing deserves more attention than it gets in the referential signaling literature