Titi semantics: Context and meaning in Titi monkey call sequences

Philippe Schlenker, Emmanuel Chemla, Cristiane Cäsar, Robin Ryder, Klaus ZuberbühlerNatural 2016 · DOI

TL;DR

This paper applies formal linguistic semantics to analyze the alarm calls of Titi monkeys, arguing that their complex sequences are not evidence of a sophisticated syntax but rather arise from simple call meanings interacting with environmental context and competition rules. This approach offers a more parsimonious explanation for Titi monkey communication than previous theories.

What Problem Does This Paper Address?

The study of animal communication, particularly primate alarm calls, has long been interested in whether these vocalizations exhibit "linguistic" properties like syntax and semantics. Previous research on Titi monkeys ( Callicebus nigrifrons) suggested that their call sequences, composed of just two basic call types (A and B), could convey complex information about predator type and location, hinting at a sophisticated interface between syntax and semantics. However, a challenge arises when trying to assign simple, consistent meanings to these individual calls, as both A and B calls appear in a wide variety of contexts, making it difficult to pin down their specific semantic content. This paper aims to resolve this challenge by applying contemporary semantic tools to re-analyze the Titi monkey call data, seeking a more explanatory account of their communication system.

Methods

The authors employ formal semantic methods, drawing parallels to linguistic analysis of human language. They focus on three key components:

  • Semantics (Literal Meaning): Assigning basic, context-independent meanings to individual calls.
  • Pragmatics (Competition Rules): Modeling how calls compete with each other based on informativity, similar to scalar implicatures in human language. This suggests that a more specific call might be preferred over a general one.
  • World Knowledge/Environmental Context: Incorporating information about the external environment and how it influences the interpretation of calls.

They propose a "null theory" of Titi call interpretation, assuming calls are produced and interpreted individually, with meanings relativized to the precise time of utterance. This allows for the environmental context to change during a sequence, influencing the interpretation of subsequent calls.

Datasets

The study primarily analyzes data from Cäsar et al. (2013), which includes:

  • Model Predator Experiments: Recordings of Titi monkey vocalizations in response to models of raptors and cats placed in different locations (canopy or ground).
  • Naturalistic Observations: Recordings of Titi monkey vocalizations in response to naturally occurring events like flying raptors, perched raptors, capuchins, and non-predator-related situations (foraging, descending, feeding).

The paper also references data from other primate studies (e.g., Campbell's monkeys, Putty-nosed monkeys) for comparative purposes.

Key Insights and Ideas

  • No Complex Syntax: The paper argues against the idea that Titi monkey call sequences possess a complex syntax or a sophisticated syntax/semantics interface. Instead, the complexity arises from the interaction of simpler elements.
  • Simple Call Meanings: The core insight is that individual calls (A and B) have relatively simple, underspecified meanings. The B-call is a general "noteworthy event" alert, while the A-call signals a "serious non-ground alert."
  • Context is King: The rich meaning of Titi call sequences is largely derived from the environmental context at the time of each call's utterance. The interpretation of a call is not just about what it directly signifies but also about how that meaning interacts with the ongoing situation.
  • Competition and Enrichment: Rules of competition, similar to scalar implicatures, play a crucial role. A more informative call (like A) can "outcompete" a less informative one (like B) when applicable, leading to an enriched meaning for the less informative call when the more specific one is absent. This explains why B-calls are not used in all situations where a noteworthy event occurs.
  • Time-Relativized Meaning: The meaning of each call is tied to the specific moment it is uttered. This allows for sequences to reflect changes in the threat level or nature over time, preventing contradictions.

Key Results

  • The authors provide a semantic specification for the B-call as applicable to any "noteworthy event" and for the A-call as a "serious non-ground alert."
  • They demonstrate how these simple meanings, combined with pragmatic competition rules and context-dependent interpretation, can account for the observed Titi monkey call sequences in various predator and non-predator situations.
  • For example, the A++ sequence in "raptor in the canopy" reflects a serious non-ground threat, while the A++B++ sequence in "raptor on the ground" suggests an initial non-ground alert that diminishes as the threat is perceived as less immediate.
  • The paper shows that the complexity observed in Titi call sequences is not due to a complex linguistic system but rather to the dynamic interplay of simple semantic components and the environment.

Limitations and Future Work

  • Limited Scope: The analysis focuses primarily on A and B calls and does not deeply explore other call types, such as the "C-call," which might be related to group movement.
  • Cognitive Underpinnings: While the pragmatic competition mechanism is analogous to human implicatures, the paper cautiously avoids claiming that Titi monkeys possess complex "theory of mind" abilities. Further research is needed to explore the cognitive and neural mechanisms behind these rules.
  • Further Validation: The authors acknowledge that some assumptions, particularly regarding the "serious" component of the A-call and the specific environmental motivations, require further empirical assessment.
  • Cross-Species Comparison: The study highlights the need for a detailed typology of primate languages to understand the evolution of communication. Future work could explore similar analyses in other primate species.
  • Experimental Modulation: The authors suggest future studies could investigate how lexical meanings are acquired and modulated in animals through experimental manipulation, similar to research on dogs and parrots.

Summary updated

BibTeX
@inproceedings{schlenker2016titi,
  author = {Philippe Schlenker and Emmanuel Chemla and Cristiane Cäsar and Robin Ryder and Klaus Zuberbühler},
  title = {{Titi semantics: Context and meaning in Titi monkey call sequences}},
  booktitle = {Natural},
  year = {2016},
  doi = {10.1007/s11049-016-9337-9},
}
Figure