Recommended Free Tools
Yes—but only in a carefully limited sense. A 2025 study of wild bonobos found that three naturally occurring call combinations had inferred meanings that were not simply the sum of their component calls. That is evidence of nontrivial compositionality, a language-like property in which meaningful elements combine to create a distinct larger meaning.
It is not proof that bonobos speak a human language, use human-like grammar, or have been “decoded.” The researchers analyzed context and statistical patterns, not bonobo-to-English translations.
As an Amazon Associate I earn from qualifying purchases.
The finding, in plain English
Wild bonobos appear to combine calls in ways that can produce a meaning different from simply adding the meanings of the individual calls. That is a significant parallel with one important property of human language: compositionality.
It is not evidence that bonobos speak a human language, use grammar like ours, or have suddenly been “translated.” The 2025 study found statistical evidence consistent with nontrivial compositionality in three naturally occurring call combinations. The researchers inferred those combinations’ meanings from the situations in which the calls occurred; they did not assign bonobo calls reliable word-for-word English equivalents.
#1 Best Overall
The distinction matters. The strongest defensible claim is that bonobo vocal communication may contain a language-like building block that scientists once expected to be unusually rare outside humans.
What compositionality means
Human speakers routinely build larger meanings from smaller meaningful parts. A listener who knows the words “rain” and “coat” can understand raincoat, even if that compound refers to a particular kind of garment rather than to rain and a coat considered separately. In other expressions, one element changes or narrows the interpretation of another.
That general principle is called compositionality. Trivial compositionality would mean that a combination’s meaning is essentially the sum of its parts. Nontrivial compositionality is more interesting: the components contribute meaning, but the combination also has a modified or emergent interpretation that cannot be predicted by simply placing the component meanings side by side.
The bonobo study did not show that a call is equivalent to a human word. It asked a narrower and more testable question: when bonobos produce certain call combinations, does the combination occupy a meaningful contextual position related to its component calls, while also differing from a simple additive prediction?
How the bonobo study was carried out
The peer-reviewed study by Mélissa Berthet, Martin Surbeck, and Simon W. Townsend was published online in Science on April 3, 2025, and in the journal’s print issue on April 4. The observations came from wild bonobos (Pan paniscus) in the Kokolopori Bonobo Reserve in the Democratic Republic of the Congo.
Study at a glance
- Field site: Kokolopori Bonobo Reserve, Democratic Republic of the Congo
- Recording effort: approximately 400 hours
- Single calls analyzed: 567
- Call pairs analyzed: 425
- Contextual features documented: 336 for each vocalization
- Repertoire: seven principal call types used in 19 observed combinations
- Core result: four combinations met the study’s compositionality criteria; three were nontrivially compositional
Rather than treating each call as if it had one fixed dictionary definition, the researchers recorded a large set of contextual details. Those details included the social and physical circumstances surrounding a vocalization, such as what the animals were doing, who was nearby, and what was happening immediately before or after the call.
The analysis used tools borrowed from distributional semantics. In simplified terms, a vocalization was represented by the contexts in which it appeared. Individual calls and combinations were placed in a multidimensional “semantic space.” The researchers then tested two relationships:
- Whether a combination was meaningfully related to its component calls rather than to unrelated calls.
- Whether the combination’s inferred meaning differed from what would be expected by simply adding the inferred meanings of its components.
This is a sophisticated way to analyze a graded, context-sensitive communication system. It is also why the results should be described as inferred meanings or contextual interpretations, not translations. The analyzed dataset and statistical code were made available by the authors through a figshare repository, allowing other researchers to inspect and build on the analysis.
The four combinations that stood out
The researchers identified four combinations that met their criteria for compositionality. Three were classified as nontrivial; one was more consistent with an additive interpretation.
| Call combination | Classification | Contextual interpretation |
|---|---|---|
High-hoot_Low-hoot |
Nontrivial | Potentially recruiting attention during distress or interrupting another individual’s display |
Peep_Whistle |
Nontrivial | Associated with tense social situations, including sexual- or dominance-related interactions in reported contexts |
Peep-yelp_High-hoot |
Nontrivial | Associated with group encounters and nest-building contexts |
Yelp_Grunt |
Trivial | More consistent with an additive combination of the component-call meanings |
The underscores indicate call combinations as represented in the study. The labels are not English translations. They identify acoustic call types, and the descriptions in the final column are hypotheses based on contextual distributions.
What the individual calls may do
The component calls also appeared in different recurring circumstances:
- Grunts occurred during grooming, resting, moving, and feeding. They may help coordinate attention during ordinary social activity.
- High-hoots appeared to signal a caller’s presence or location, particularly over distance or in potentially dangerous situations.
- Low-hoots were associated with high-arousal situations, including night-nest construction.
- Peeps and yelps both helped coordinate social activity. The researchers’ tentative interpretation was that peeps function more like suggestions, while yelps may act more like stronger imperatives.
Words such as “suggestion” and “imperative” are useful shorthand for patterns in the data, but they should not be mistaken for proof that bonobos possess human-like grammatical moods or consciously formulate commands in the same way people do.
Why the combinations are more than a call list
Suppose every call had only one fixed signal function and every pair simply repeated both functions. That would still be interesting, but it would not tell us much about a system capable of building richer meanings.
In the three nontrivially compositional combinations, the combination was related to its constituent calls but was not adequately described as their simple sum. For example, High-hoot_Low-hoot was not merely “the high-hoot meaning plus the low-hoot meaning.” Its contextual profile suggested a distinct social use involving attention, distress, or interruption. The same general pattern applied to Peep_Whistle and Peep-yelp_High-hoot.
The study does not establish exactly which component modifies which, whether the order of calls carries a stable grammatical function, or whether bonobos can apply the same pattern freely in other combinations. It establishes that the combinations have a structure worth investigating.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Why this is a major result in animal communication
Many animals have rich communication systems. Some signals vary with context, some are combined, and some appear to refer to particular predators, foods, individuals, or situations. But those observations do not automatically demonstrate compositionality.
The importance of this study is its scope and method. The researchers did not focus on one famous alarm call or a single unusual sequence. They examined a substantial portion of the observed vocal repertoire and modeled the calls in relation to many contextual features. Their result suggests that compositionality may be a prominent feature of bonobo vocal communication rather than an isolated exception.
That pushes the comparison with human language beyond the simple question of whether animals can make two sounds in a row. The more consequential question is whether a communication system contains meaningful parts that can interact to create a distinct whole. On that narrower question, the evidence from bonobos is unusually strong compared with much previous work on wild animals.
Rank #3
What the study does not show
Several popular interpretations go beyond the evidence. The study did not demonstrate any of the following:
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11- That bonobos have human language.
- That researchers decoded a complete bonobo dictionary.
- That every bonobo call has one stable English equivalent.
- That bonobos use human-like syntax or hierarchical grammar.
- That they use recursion, displacement, or a large open-ended vocabulary.
- That they can create unlimited novel sentences or call sequences.
- That they intentionally communicate abstract propositions in the human sense.
- That other bonobos will reliably understand any unfamiliar combination.
The study primarily analyzed combinations that occurred naturally. That is essential evidence, but it leaves a central question unanswered: Can bonobos generalize the pattern?
For example, if researchers observed a previously undocumented combination, would listeners respond in a way predicted by the meanings of its components? Could bonobos produce or understand unfamiliar combinations in appropriate contexts? A positive answer would provide much stronger evidence of a productive compositional system.
Johan Bolhuis, a neurobiologist at Utrecht University, made a related criticism in contemporaneous reporting: compositionality should not be reduced to the mere fact that two calls or “words” can be combined. Human language also involves syntax and larger, structured units of meaning. The study’s authors likewise acknowledged that the relationship between these combinations and the rest of the bonobos’ signaling system remains unknown.
Why the method matters—and where it can be cautious
One of the study’s most useful contributions is methodological. Researchers often want to ask what an animal signal “means,” but animal signals rarely behave like entries in a bilingual dictionary. A call may be used across several contexts, and the listener’s response may depend on the caller, audience, social relationship, location, arousal level, and immediate event.
Free tools Windows power users keep installed
One-click scans. No signup required.
Distributional analysis accommodates that complexity. Instead of forcing a call into one label such as “danger” or “come here,” it looks for a pattern across many observations. That makes the approach better suited to communication systems whose meanings are probabilistic and context-dependent.
It also means the conclusions depend on the quality and range of the observations, the way contextual features were defined, and the statistical assumptions used to compare meanings. A semantic-space result is not a recording of a bonobo mentally stating an English sentence. It is evidence that particular vocal patterns are distributed through social situations in a structured way.
A further complication is that call categories themselves may not be as cleanly separated as their labels suggest. A separate 2025 paper describing an updated vocal repertoire of wild adult bonobos emphasized how difficult it is to resolve vocal repertoires in the field and suggested that bonobo communication may be more graded and complex than a set of sharply discrete call types implies. Future analyses could therefore revise the number or boundaries of the calls used in compositionality research.
What this may mean for the evolution of language
Bonobos and chimpanzees are humans’ closest living relatives. The bonobo–chimpanzee lineage split after the lineage leading to humans had already diverged, so bonobos are not ancestral humans and cannot provide a recording of our own linguistic past.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsRank #4
They can, however, help researchers identify which communication capacities may have deep evolutionary roots. When this study is considered alongside earlier evidence that chimpanzees combine calls, it raises the possibility that some capacity for compositional communication existed in a common ancestor of humans, bonobos, and chimpanzees.
That is an evolutionary hypothesis, not a demonstrated date on the history of language. The data cannot reveal precisely when compositionality evolved, whether it arose once or multiple times, or how it may have interacted with changes in cognition, social life, anatomy, and learning. The study narrows the apparent gap between human and other-ape communication, but it does not erase the many differences between them.
The most exciting implication is therefore not that bonobos are “almost speaking.” It is that a property once treated as especially characteristic of human language may have a deeper and more distributed evolutionary history than researchers assumed.
What should researchers test next?
The next step is controlled testing rather than more imaginative translation. Several experiments would be especially valuable:
- Playback of known combinations: Play recordings to bonobos who did not produce them and measure whether their attention, movement, affiliation, alarm, or other behavior changes in the predicted way.
- Novel combinations: Present unfamiliar sequences assembled from known call types. If listeners respond according to a compositional rule, that would be stronger evidence than an association found only in naturally occurring examples.
- Order tests: Compare the same calls in different orders to determine whether sequence position matters or whether the calls simply accumulate.
- Cross-group replication: Test whether the patterns hold in other wild communities rather than only in Kokolopori.
- Developmental studies: Examine whether young bonobos learn these combinations, whether usage changes with age, and how much variation is shaped by social experience.
- Cross-setting comparisons: Compare wild, captive, sanctuary, and human-raised bonobos while controlling for differences in environment and social structure.
As of August 12, 2026, the closest directly relevant follow-up identified in the searched literature was a 2026 Evolang conference presentation on vocal plasticity in bonobos. It compares animals from wild, captive, sanctuary, and human-raised settings and examines differences in vocal production and usage. The presentation is not a completed peer-reviewed replication of the 2025 compositionality finding, so it should be treated as an active research direction rather than confirmation.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Two bonobo research stories that should not be confused
Readers may encounter famous accounts of bonobos trained to use symbols or lexigrams, including the language-trained bonobo Kanzi. Those studies are relevant to the history of great-ape cognition, but they are different from the 2025 research. Kanzi-related work concerns communication in a captive, human-mediated setting; Berthet, Surbeck, and Townsend studied naturally occurring vocal combinations among wild bonobos.
Neither line of research should be used to stand in for the other. A captive ape’s performance with human-designed symbols does not prove that wild bonobos use the same system, and wild call compositionality does not depend on human training.
For readers who want broader background on bonobo behavior and its implications for human evolution, Bonobo: The Forgotten Ape by Frans de Waal and Frans Lanting is a natural companion read. It provides context about bonobos; it is not a source for the 2025 call-combination result. Readers interested in the history of great-ape language research can also look for Kanzi: The Ape at the Brink of the Human Mind, while keeping its captive, human-mediated subject matter distinct from this wild-bonobo study.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Disclosure: Book links on this page may be affiliate links. That does not change the distinction between background reading and evidence from the study itself.
Best Value
What this means for people who live with pets
This finding does not provide a decoder for dog barks, cat vocalizations, or other companion-animal sounds. Communication systems are species-specific, and a call’s apparent meaning depends heavily on the social and physical context in which it is produced.
It does offer a useful caution against two opposite mistakes. Animals are not simply making meaningless noises, but neither should every familiar sound be translated into a human sentence. The bonobo evidence supports studying animal communication as structured, flexible, and context-sensitive without assuming that it is human language in disguise.
The careful conclusion
Bonobos may have one of the strongest known examples of language-like structure in a wild animal communication system. Three combinations—High-hoot_Low-hoot, Peep_Whistle, and Peep-yelp_High-hoot—showed evidence of nontrivial compositionality, while Yelp_Grunt fit a more straightforward additive pattern.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →That is a genuine advance. But it is evidence for a specific property of communication, not proof that bonobos have human language. The decisive future test will be whether bonobos can understand and use novel combinations in predictable ways across contexts and groups. Until then, the most accurate headline remains the intriguing one: bonobo calls may be the closest thing to animal language scientists have yet found—not because bonobos speak like people, but because their calls appear to combine meaning in a surprisingly structured way.
Frequently Asked Questions
Do bonobos have language?
No. The study found evidence for one language-like property—nontrivial compositionality—in three naturally occurring call combinations. It did not demonstrate human-like grammar, recursion, a large open-ended vocabulary, or the ability to create unlimited novel sentences.
What is nontrivial compositionality?
It means that the meaning inferred for a call combination was not adequately explained by simply adding the inferred meanings of its component calls. The combination was related to its parts but also had a distinct contextual profile.
Did scientists translate bonobo calls?
No. Researchers inferred contextual meanings by comparing calls with hundreds of recorded features and the situations in which they occurred. The study did not produce a complete dictionary or word-for-word English translations.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →How can scientists prove that bonobos understand novel call combinations?
Researchers need playback experiments, tests with unfamiliar combinations, comparisons of different call orders, and replication across bonobo groups. Those tests can show whether bonobos generalize compositional patterns beyond the combinations already observed.
The Bottom Line
Bottom line: The 2025 study found evidence that three natural bonobo call combinations have non-additive, compositionally structured meanings. That brings wild bonobo communication closer to one important property of human language, but it does not establish human-like grammar, limitless productivity, or a translatable bonobo vocabulary.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Leave a Reply