The “Language” of Honeybees — A Revision
2025年9月4日
Jürgen Tautz
For nearly a century, the “dance language” of honeybees—first decoded by Nobel laureate Karl von Frisch—has captivated scientists and sparked fierce debate. But new research is challenging the classic view, revealing that the secret to bee communication is far more complex than a simple waggle. Discover how fresh insights from a major reference work are reshaping our understanding of one of nature’s most celebrated animal behaviors.
In 1973, the Austrian zoologist Karl von Frisch received the Nobel Prize for his studies on honeybee behavior, including his observations and ideas on how bees communicate with each other. https://www.nobelprize.org/prizes/medicine/1973/frisch/facts/
The central idea — often reiterated even today (“only humans and honeybees can communicate in an abstract manner the precise data on the coordinates of a target”) — has been regarded as one of the premier scientific hypotheses of the 20th century (LA Times, 1991). https://www.latimes.com/archives/la-xpm-1991-11-05-vw-842-story.html
This concept excited not only biologists but also linguists, psychologists, and others. The widespread interest in this topic, its history, and the often fierce debates it has elicited, motivate an update in this area for the Encyclopedia of Language and Linguistics.
Here, the relevant aspect is not whether such a form of communication qualifies as a language in the linguistic sense (see the respective contributions in ELL 2026), but rather to show that what is called “danced language in honeybees” is not an abstract and precise transmission of information. Instead, what appears to originate from a “language” is actually the result of a sequential series of behavioral signals and cues.
Before embarking on their next foraging trip, worker honeybees returning from successful foraging perform a peculiar stereotypical movement pattern to communicate the location of a food source within the darkness of their nest, on the surface of the vertical hanging combs. This behavior has been known since Aristotle’s time, but it was not until Karl von Frisch’s investigations that it was understood to follow a regular pattern. He called this the waggle dance (Fig. 1).
Fig. 1: The “von Frisch phase” of honeybee recruitment (phase 1 in the new model) begins with the waggle dance (the blurred bee in the upper half of the photo) inside the darkness of the nest. (copyright Ingo Arndt)
The choreography of the waggle dance varies with the position of the sun or, at the same time of day, with the relative position of the target to the hive. The dance also encodes the distance to the target: recruited bees that have observed a dance can locate the advertised target, even if it is as small as a saucer 10 km away.
Von Frisch’s discovery of the regularity in the dance, combined with the reliable arrival of recruits at the target, led him to conclude that bees find the target by following the information contained in the dance. This marked the birth of the so-called ‘dance language’ of honeybees.
This interpretation elevated honey bees above most other animals in terms of communication. As an example of the reaction it provoked: “Only human beings and honeybees possess the ability to guide others to an important location by providing abstract information — its direction and distance from the current location.” This element was decisive in awarding von Frisch the Nobel Prize in 1973.
Since the publication of the dance language recruitment theory nearly 80 years ago, one problem has persisted from the beginning and has significantly influenced subsequent research: the dances advertising a specific target, the ‘message’ to the recruited bees, show considerable variability, yet recruits still manage to arrive at the advertised spot. To address this issue, several auxiliary hypotheses have been proposed.
Auxiliary hypotheses
The first hypothesis, proposed by von Frisch himself and accepted as correct to this day, suggests that dance followers calculate the arithmetic mean from observed dances. Based on the directional information of the waggle phases, estimated by human observers, the true direction to the target can be derived. The accuracy of this calculation depends on the number of observations and involves omitting the first and last dance rounds in multi-circle dances. This mathematical explanation relies on the data set showing symmetry around the true direction.
A second hypothesis assumes that honeybees possess a cognitive map of the landscape, enabling them to perform vector calculations to locate the target.
A third hypothesis posits that recruits are not necessarily reaching the exact advertised target but are instead distributed over an area around it — a strategy that benefits the colony’s foraging efficiency.
A fourth hypothesis suggests that, on their first flight, bees follow the route and altitude taken by the dancer. Bees estimate flight distance based on optic flow — the perceived movement of objects passing by during flight — which depends on the environment encountered along the way.
A fifth hypothesis, inspired by observations that scent alone can guide recruits to the target, claims that honeybees have multiple cues available to locate the goal. According to this view, the waggle dance is just one of several methods bees use to find food. Proponents of this extreme position deny that any geographical information in the dance is used by followers.
Lack of definitive evidence
Each of these hypotheses has supporters, but none is supported by conclusive data. Over the decades, this has led to a complex and sometimes contradictory picture. Nevertheless, the classical interpretation — that honeybee communication relies solely on the waggle dance — has remained dominant.
This explanation has been repeatedly reiterated in scientific studies, textbooks, and publications. For example: “By performing this dance, successful foragers share information about the direction and distance to patches of flowers yielding nectar and pollen, water sources, or new nest sites with other colony members.”
A new model, presented by J. Tautz and D. C. Sandeman in ELL 2026, bypasses the need for auxiliary hypotheses and bridges the divide that has fueled a decades-long “bee battle” over the cues and signals used by recruits. The new model describes recruitment as a three-phase process:
Phase 1: Sent — The recruit flies out of the hive in the approximate direction and distance indicated by the dance, heading toward the general search area.
Phase 2: Search — The bee reaches the search area, where the exact location, shape, and extent of the search zone are influenced by external factors (such as weather) and internal factors (such as motivation).
Phase 3: Attraction — If the recruit encounters floral scent or signals from the dancer in the field, it can orient toward and arrive at the target.
In this model, the function of the dance is to guide followers into a broad area where they are likely to encounter cues that lead them to the goal. The first phase, here called the ‘von Frisch phase,’ involves the initial directional guidance. The final phase, here called the ‘Wenner phase,’ is dominated by olfactory cues, as studied by Adrian Wenner and colleagues, involving scent-guided target finding.
In ELL, Tautz and Sandeman include a graph comparing the old and new models. This visual aid has been widely adopted in educational settings, as the “dance language” in honeybees remains one of the most popular and studied animal behaviors, and the update has been well received.
Not a single observation or fact supports the old model better than the new one — in fact, numerous pieces of evidence favor the latter. For example, the first flight to the target by newly recruited bees can take up to 50 times longer than the dancer’s own travel time. Additionally, new recruits carry three to four times more honey than dancers do before their initial foraging flights. Interestingly, von Frisch’s early explanation for the precise arrival of recruits was based solely on Phase 3 of the new model. He also discovered that the same bees performing waggle dances in the hive produce loud buzzing flights with open Nasonov glands that release attraction pheromones in the field (Fig. 2). This later led him to overlook the interaction between bees at the target, where strong pheromone signals are also emitted.
It is fascinating that this three-phase strategy for reaching distant goals appears in other contexts, such as long-distance migration in birds or space missions like the recent DART project targeting the 160-meter-diameter asteroid 11 million kilometers away.
Fig. 2: The “Wenner phase” of honeybee recruitment (phase 3 in the new model) involves the opening of the Nasonov gland (the yellowish band on the abdomen) and the release of an attraction pheromone that guides recruits toward the target. As von Frisch discovered, forager bees that perform the waggle dance in the hive continue this recruitment process in the field, using scent signals. (copyright Ingo Arndt)
貢獻者
JT