The Archived Record: My Communication Hypothesis Was Public in 2012
The historical record matters because these ideas are not a recent response to the Boxgrove research. My Revelation in Stone website was publicly online by at least July 2012, and my own contemporaneous account indicates that I had begun finding what I interpreted as prehistoric artworks containing a “hidden language element” during the previous year, in 2011.
On 29 July 2012, while linking directly to my Revelation in Stone website, I publicly described the artefacts as having a “hidden language element” and stated that geographically separated material from the USA appeared to contain the same glyphs. Within days, on 2 August 2012, I described the subject simply as “Flint artwork language.”
The surviving 2013 website and blog material preserves the hypothesis in considerably greater detail. By then I was explicitly describing worked stone artefacts and flint tools as “visual language devices”, “prehistoric language tools” and “communication tools”, while discussing repeated visual “words”, recurring iconography, animation, multi-image or poly-glyph artefacts, and practical silent communication concerning animals and hunting.
The important point is therefore simple:
The language-and-glyph hypothesis was already public in 2012, and my own dated statement places its beginnings in 2011. The surviving 2013 archive does not mark the beginning of the idea; it preserves a more detailed version of a communication model I was already publicly presenting.
The terminology is also significant. I was approaching worked prehistoric stone explicitly in linguistic terms - language, glyphs, visual words, communication and lexicon-like recurring vocabulary - long before the recent Shipton et al. Boxgrove paper presented its own independently derived argument for lexical specificity in Lower Palaeolithic stone-tool culture.
Archived Revelation in Stone homepage 24 October 2013
Research Published After My 2012 Language and Glyph Work
When I began publicly describing prehistoric worked flint in terms of a hidden language, recurring glyphs and visual communication in 2012, these ideas sat far outside the archaeological interpretation normally applied to stone tools.
Since then, a substantial body of published research has increasingly explored relationships between stone technology, symbolic communication, language, conventional signs and information carried by portable prehistoric objects.
2013 — Uomini & Meyer: Acheulean toolmaking and language show closely correlated patterns of brain lateralisation.
2015 — Morgan et al.: experimental transmission of knapping improves with teaching and particularly language; teaching or proto-language is proposed as potentially important to Acheulean technology.
2017 — Lombao et al.: gesture and verbal communication improve the teaching of stone-tool production, with especially strong results from verbal teaching.
2018 — Majkić et al.: an engraved Neanderthal-associated flint is considered capable of having been deliberately marked to express or communicate meaning.
2020/2021 — Barham & Everett: Lower Palaeolithic stone technology is examined as material evidence for symbols, semiotics and early language.
2021 — Etxepare & Irurtzun: selected Palaeolithic hand configurations are investigated as possible sign-language units.
2026 — Bentz & Dutkiewicz: thousands of signs on portable Aurignacian objects are analysed as an intentional and conventional sign system.
These papers do not collectively prove my proposed decoding, and chronology alone cannot tell us whether any particular researcher encountered my work. What the publication record does demonstrate is that many concepts closely relevant to the framework I was publicly discussing in 2012 — language associated with stone technology, conventional signs, portable information-bearing objects, visual communication and lexicon-like systems — have subsequently become serious subjects of published archaeological and experimental research.
It is increasingly possible to ask not simply whether prehistoric people possessed sophisticated communication, but what traces of those communication systems may remain physically preserved in their tools, signs and portable objects.
An Essential Qualification: Not One Time, One People or One Function
An important qualification is necessary when discussing this system. I do not know the individual age of every object, and I have never assumed that the wider material represents one moment in prehistory, one population or one continuous archaeological culture. Different stones may represent different occupations, traditions and periods. Two apparently related objects could be separated by thousands, hundreds of thousands or potentially much greater spans of time. Similarity between them therefore has to be investigated as recurrence of convention, not automatically treated as proof of contemporaneity. It certainly appears as though one species used the same stone art tradition through vast periods of time. (or could be interpreted thus)
Nor should the maker automatically be assigned to Homo sapiens. In deep contexts the responsible language is simply a hominin or human-related maker until better evidence identifies the population. Different object types could also have belonged to different users: hunters, knappers, teachers, children, storytellers, ritual specialists or social roles for which there is no modern equivalent.
The research question is therefore not “what single language did all these stones speak?” It is: which visual conventions recur, how are they physically organised, and do they repeatedly perform the same communicative operations?
Not Every Stone Is a Sentence: Different Communication Devices
Some objects appear simple. Others are densely structured. That variation is not a weakness in the hypothesis; it may be exactly what we should expect from material communication. A warning sign carries less information than a diagram, but both can belong to the same communication culture.
The following device classes summarise several recurring types I have identified. Their proposed meanings remain hypotheses, but the operations themselves, selection, rotation, reversal, animation and recurring placement, can be demonstrated directly from the objects.
Ape Face: Subject + Gaze + State
The grey ape-face find is one of the clearest examples of the higher-information model. I interpret several possible eye or pupil forms within the socket areas, together with mouth/tongue, ear and other facial details that are unnecessary merely to make the object recognisable as an ape.
- Recognise the ape subject.
- Select or point to an eye/pupil to indicate gaze direction, awake, sleeping or attention.
- Rotate the stone to indicate the direction the animal or head is facing.
- Point to mouth/tongue as a possible feeding state.
- Point to a closed eye as possible rest/sleep.
- Point to the ear as possible hearing or alertness.
Flat Blade / Scraper: Rotate to Select the Species
A recurring flat blade or scraper type carries several face profiles worked along its edges. In my interpretation these commonly concentrate on apes — baboon, chimpanzee, gorilla and related profiles. The stone can be rotated until the intended profile is uppermost.
- Rotate the tool to expose the required animal profile.
- The uppermost profile establishes the species or animal type.
- Its facing direction supplies directional information.
- Flip the blade and a corresponding animal can face or travel the opposite way.
Simple Animated Figure: Action Rather Than Dense Encoding
Not every object appears to carry multiple lexical details. Some simple cloaked or human-like figures can be animated in the hand so that changing orientation produces the impression of walking.
Such an object might do no more than convey a person, movement or direction — or it may simply have been made for animation, teaching, storytelling or amusement. The point is that a communication system can contain very simple units as well as densely encoded ones.
Elephant Composite: A More Conceptual Object
Another recurring type is more difficult to reduce to immediate hunting information. In its basic form I recognise an elephant, a genital/vulval sign placed between trunk and front leg, and an associated ape face. Many examples contain considerably more imagery than this basic recognition.
If intentional and conventional, such a combination may concern sex, fertility, reproduction, identity, classification, narrative or some other relationship between the represented concepts. It is precisely the kind of object that warns against forcing every artefact into the “silent hunting aid” explanation.
Rotational and Multi-image Devices
Some objects change subject when turned. A line or flake scar can belong to one animal in one orientation and another in the next. This allows the physical manipulation of the object to become part of the reading.
If deliberate, rotation is not decoration; it is an operation. One artefact can hold several subjects, directions or relationships without requiring a separate object for each.
Dinner and Danger: The Most Basic Survival Logic
My original “dinner and danger” explanation remains one of the simplest ways to understand why animal information could matter. In a prehistoric landscape an animal may represent food, competition or immediate physical danger. The basic questions are brutally practical:
Imagine several people hearing a dangerous animal nearby. Calling out its identity may expose the group. A shared visual referent could establish the animal silently; orientation and selected details could add what is known about its direction or state.
Likewise, the heavy concentration of ape-like profiles in some tool types may indicate that apes were highly salient to the makers — as prey, competitors, threats, objects of close observation, culturally important animals, or some combination of these. The imagery alone cannot tell us which role dominated, but repeated species recognition itself would be meaningful.
The Visual Device Does Not Replace Language It May Depend on It
A visual communication device does not require a speechless population. It may imply the opposite. A conventional system first has to be learned. Someone has to understand which feature matters, what a particular orientation means, and how another person is expected to read it.
Experimental archaeology supports the importance of active communication in stone-tool teaching. Morgan and colleagues compared different ways of transmitting Oldowan knapping and found that teaching improved transmission, particularly verbal teaching. They concluded that “teaching or proto-language may have been pre-requisites” for Acheulean technology.
Later experiments similarly found that gesture and verbal communication can outperform observation alone in learning structured knapping procedures. Verbal labels are especially useful for abstract or difficult-to-see relationships such as platform selection, angle and the sequence connecting one removal to the next.
Lombao, Guardiola & Mosquera (2017), Teaching to make stone tools
THE STONE CARRIES THE CODE.
GESTURE SELECTS THE INFORMATION.
THE MESSAGE CAN THEN BE TRANSMITTED SILENTLY.
This produces a coherent survival model. The conventions could be explained around camp through speech, demonstration and repeated use. During hunting, stalking or danger, the same conventions could then be selected without vocalising. Speech and material communication are therefore complementary, not competing explanations.
Strong Evidence Consistent With Verbal or Proto-verbal Language?
The cumulative argument becomes stronger when technology and the proposed visual vocabulary are considered together. Complex knapping already requires concepts such as edge, platform, angle, thinning and sequence. If the visual system is also intentional, it adds another range of socially distinguished concepts: eye, ear, tongue, hand, thumb, animal, sex, direction, gaze, feeding, movement and other repeated subjects or states.
A surviving image does not prove the exact spoken word used for it. Nor does a conventional sign require speech in principle. But a socially shared repertoire containing technological concepts, animal categories, body parts, actions, directions and behavioural states is entirely consistent with a sophisticated verbal or proto-verbal lexicon.
Speech also solves a practical problem. Knappers and hunters often have their hands occupied by cores, hammerstones, tools, carcasses or weapons. Vocal communication allows concepts to be transmitted while the hands and eyes remain engaged. Once the visual code had been learned, however, that same code could become valuable precisely when speech was undesirable.
Shipton et al. at Boxgrove: Two Routes to Symbolic Language
The recent Boxgrove research by Shipton and colleagues addresses the same broad problem as my work, although the communication system they identify is different from my proposed pictorial lexicon. Their paper presents two related but largely independent lines of evidence leading toward symbolic language.
First, the standardized tranchet tip was a directly observable feature of the Boxgrove handaxes. Ceri Shipton describes this as a social norm: a culturally standardized feature that members of the group recognised and were expected to reproduce. In this sense, part of the shared social vocabulary was physically visible in the artefact itself.
This kind of normativity is relevant to symbolic language because a communication system depends upon members of a group sharing conventions — agreeing that a particular form, sound, gesture or sign has a recognised meaning.
The second argument concerns turning-the-edge. This is a difficult and visually opaque knapping technique whose underlying relationships may not be readily understood through observation or even pointing alone. Shipton and colleagues therefore argue that teaching such a technique may have required symbolic language and specific conventional referents — a lexicon.
These are importantly different arguments. The tranchet tip concerns a directly visible and culturally standardized feature of the handaxe. Turning-the-edge concerns the symbolic communication needed to transmit a difficult technological concept. Both lead the authors toward symbolic language.
Shipton et al. — Turning-the-edge, Tranchet, and Social Signalling at Boxgrove
My additional proposal is that some worked artefacts preserve a further pictorial lexicon: recognisable subjects, anatomical details, orientation and transformations that may themselves have carried information.
Two Lexicons in the Same Communication World?
Shipton et al.'s Boxgrove model and my proposed visual lexicon need not compete. They may describe different informational layers within material culture.
| Layer | Possible content | How it survives |
|---|---|---|
| Boxgrove social / technological vocabulary | Standardized handaxe features and socially shared technological concepts, including the tranchet form and concepts required for techniques such as turning-the-edge. | Partly visible directly in culturally standardized tool features such as the tranchet tip; partly inferred from the symbolic communication required to teach difficult technological procedures. |
| My proposed visual / referential lexicon | Elephant, ape, chimpanzee, baboon, gorilla, bear, egg, hand, thumb, person and other proposed subjects. | Proposed to remain materially visible as recurring pictorial and visual conventions incorporated into worked artefacts. |
| My proposed pragmatic modifiers | Looking, hearing, feeding, sleeping, travelling, pointing, facing this way / that way. | Proposed through selected eyes, ears, mouths, tongues, limbs, orientation, rotation and animation. |
The important clarification is that the Shipton et al. Boxgrove interpretation is not based only on an invisible or lost technological lexicon. The standardized tranchet feature itself was directly observable and formed part of a shared social vocabulary. My hypothesis goes further in proposing that recognisable pictorial subjects and modifiers were also intentionally incorporated into some artefacts.
My Proposed Lexicon Is Visible
Shipton et al.'s Boxgrove interpretation now provides an important distinction. Their tranchet-tip argument identifies a directly observable, culturally standardized feature within the handaxe as part of a shared social vocabulary, while their turning-the-edge argument points toward symbolic language and conventional referents required for technological teaching.
My own interpretation proposes an additional visible layer. I recognise recurring pictorial units directly within worked forms: elephant, ape, chimpanzee, gorilla, baboon, egg, hand, thumb, bear, human/cloaked figure and other repeated subjects — not only in my own South Downs assemblage, but also, in my interpretation, within the Boxgrove archaeological assemblage and the wider world of Palaeolithic stone tools.
I also recognise smaller units — eyes, pupils, ears, tongues, genital forms, pointing limbs, shared outlines and directional transformations — that may modify the larger subject. If such units are intentional and conventional, then they are not simply pictures. They are candidate lexical components.
The research task is to demonstrate recurrence objectively: the same unit, comparable placement, comparable operation, independent objects and, preferably, independent archaeological contexts.
Bizmoune Shell Beads: Communication With My Discovered Lexicon?
The approximately 142,000-year-old Bizmoune Cave shell beads provide an important comparison with my communication hypothesis. These were not simply attractive shells collected by prehistoric people. A highly consistent shell type was selected, deliberately modified, suspended or worn, and displayed where other people could see it. Some examples also preserve red ochre residues.
Researchers interpret the objects as durable symbolic signals: a form of nonverbal communication, probably conveying identity or social affiliation, although the precise content of that communication remains unknown.
That unknown content is what makes the Bizmoune material particularly interesting to me.
When I examine examples of these shells, I recognise visual structures that closely resemble elements of the pictorial vocabulary I had already identified independently in worked stone. These include the backward-preening waterbird, elephant, phallic form, thumb-and-finger motif, and a recurring long-necked aquatic or reptilian half-form that I have described as plesiosaur-like.
The Bizmoune researchers did not identify these figures. Their independent conclusion is that the shells were deliberately selected, modified and used as information-bearing symbolic objects, but that conclusion is exactly what I was pushing with my stone artifact and flint tool discoveries perhaps a decade earlier.
My additional observation is that some of those same communicative objects appear to contain visual components belonging to the vocabulary I have described since 2012.
Ochre, Eyes and a Restricted Visual Repertoire
The published analysis independently records red ochre residues on some Bizmoune ornaments. In several examples I observe that coloured and modified areas appear visually integrated into recurring figurative structures.
Most strikingly, some ochre spots appear positioned where they form or emphasise the eye of the backward-facing, preening waterbird motif I had already recognised elsewhere.
The potentially more important observation, however, is not any single apparent image. It is the restricted nature of the repertoire.
I do not recognise an unlimited range of arbitrary subjects in these shells. I repeatedly see a comparatively small group of forms already familiar from my earlier work:
PRENEING WATERBIRD / SWAN-LIKE FORM
ELEPHANT
PHALLIC / PENILE FORM
THUMB-AND-FINGER MOTIF
PLESIOSAUR-LIKE HALF-FORM
In some examples these motifs also occur in combinations I had previously recognised in finds from my site here in the U.K on the South Downs, near Boxgrove, particularly the ambiguous waterbird/elephant composite in which shared structure contributes to more than one reading.
This matters because a large number of random possible images would weaken the argument. A small, recurring visual vocabulary appearing repeatedly and sometimes in the same combinations is much closer to what a lexicon hypothesis would predict.
What the Bizmoune Comparison Could Mean
The comparison does not depend on claiming that the Bizmoune researchers have decoded the shells. They have not.
Their conclusion is that the ornaments formed part of a prehistoric symbolic communication system whose precise message is unknown.
My claim is narrower and separate: I recognise within some of these already-accepted communicative objects visual forms that correspond closely with motifs I had independently documented in other worked artefacts.
That creates an important testable question:
If these shells communicated something, could some of the recurring figurative structure be part of the message?
The shell itself may also have provided a useful degree of standardisation. A repeated shell type gives the maker a relatively consistent carrier. Natural curvature can be incorporated into the imagery. Deliberate modification can emphasise or complete forms. Ochre can mark selected details. Orientation can determine which part of a composite image is presented.
In principle:
STANDARD CARRIER → RECOGNISABLE IMAGE → MODIFYING DETAIL → ORIENTATION → SHARED MEANING
The exact purpose could have concerned identity, affiliation, status, memory, social information, hunting, survival or something else entirely.
What makes the Bizmoune material important to my hypothesis is the convergence:
Archaeologists independently identify the shells as prehistoric information-bearing symbolic objects.
I independently recognise within some of those objects a restricted set of visual forms already present in my proposed vocabulary.
If those correspondences prove systematic across multiple examples, they deserve direct investigation.
See my blog post: Figure Stones - Palaeolithic Language
Bizmoune Shell Beads: Accepted Communication Evidence and My Proposed Visual Vocabulary
The Bizmoune shell ornaments are already interpreted archaeologically as deliberately selected, modified and socially visible symbolic objects capable of carrying information. My additional observation is that some of these same objects appear to contain a restricted set of recurring figurative forms that closely correspond with visual motifs I had identified independently in other worked artefacts.
| Question | Published Bizmoune Interpretation | My Additional Observation / Hypothesis |
|---|---|---|
| Were the shells deliberately selected? | Yes. A highly consistent type of marine snail shell was repeatedly selected for use as personal ornamentation. | The repeated use of a similar shell form may also have provided a relatively standardised visual carrier for recurring imagery. |
| Were the shells modified? | Yes. The shells show deliberate modification associated with perforation, suspension and use. | I observe that some modified areas appear visually integrated into recurring figurative structures rather than being visually unrelated to them. |
| Was pigment present? | Yes. Red ochre residues have been identified on some ornaments. | In some examples, ochre-coloured spots appear to occupy significant positions within the imagery — particularly where they seem to form or emphasise the eye of the backward-preening waterbird. |
| Were the objects communicative? | Researchers interpret the ornaments as durable, visible symbolic signals capable of communicating social information without conversation. | This independently establishes the important principle that these were information-bearing visual objects. My question is whether some of their surviving visual structure formed part of that information. |
| Is the exact message known? | No. Researchers infer symbolic or social communication, but the precise message or vocabulary has not been recovered. | I propose that part of this presently undecoded content may remain visible in recurring figurative conventions. |
| Do I recognise a restricted vocabulary? | The researchers do not identify figurative subjects of this kind. | Yes. I repeatedly recognise a comparatively restricted set of forms: preening waterbird / swan-like form, elephant, phallic form, thumb-and-finger motif, and a plesiosaur-like half-form. |
| Do motifs recur in combinations? | No such figurative interpretation is proposed by the researchers. | Yes. Of particular interest is the recurring waterbird / elephant composite, where shared structure appears capable of contributing to more than one recognisable image. |
| Why is a restricted repertoire important? | The archaeological interpretation concerns deliberate ornamentation and symbolic signalling rather than my proposed figurative lexicon. | A limited group of recurring motifs is potentially more significant than finding unlimited possible pictures. A restricted vocabulary, especially when recurring in similar combinations, is what a lexicon hypothesis would predict. |
| Could orientation matter? | The ornaments were worn or suspended and therefore presented visually to other people. Their exact orientation-related meaning is unknown. | I propose that orientation may influence which component of a composite image becomes prominent, just as rotation and orientation operate in some of my proposed stone communication devices. |
| Does this prove my decoding? | No. The published research does not identify my proposed subjects. | No. The significance is the convergence: objects already independently interpreted as prehistoric symbolic communication signals appear, in my analysis, to contain visual forms belonging to a vocabulary I had identified independently elsewhere. |
The question is whether a restricted and repeatedly combined visual vocabulary can be demonstrated within objects that archaeology already interprets as deliberately selected, modified and information-bearing.
A large number of possible images would weaken the argument. A small, recurring vocabulary, especially appearing in the same combinations, is exactly what a visual lexicon hypothesis predicts.
2020/2021 — Stone Tools, Symbols and the Origin of Language
A particularly relevant development came from archaeologist Lawrence Barham and linguist Daniel Everett in their study Semiotics and the Origin of Language in the Lower Palaeolithic, first published online in 2020 and appearing in the 2021 journal volume.
Their approach is important because they argue that the archaeological search for language should not depend upon recovering speech itself. Instead, material culture can preserve evidence for socially understood signs, symbols and conventions. They examine Lower Palaeolithic technology in explicitly semiotic terms and argue that Homo erectus possessed at least a symbol-based form of language capable of transmitting complex cultural knowledge.
Of particular relevance to my work, they consider whether conventional stone-tool forms could possess communicative or semiotic functions as well as practical technological ones.
This research was published years after I was publicly describing worked flint artefacts in terms of language, glyphs and recurring visual communication. The hypotheses are not identical. Barham and Everett are primarily inferring symbolic communication from technological and social convention. I argue that some worked artefacts additionally preserve a recurring visible pictorial vocabulary within the objects themselves.
My additional question is whether some of the conventional signs and referents remain visibly embodied in the artefact itself.
Barham & Everett — Semiotics and the Origin of Language in the Lower Palaeolithic
2026 — A Conventional Sign System on Portable Palaeolithic Artefacts
In 2026 Christian Bentz and Ewa Dutkiewicz published a statistical analysis of Aurignacian mobile artefacts carrying deliberately engraved geometric signs dating to approximately 40,000 years ago.
Their study analysed more than 3,000 signs on 260 portable objects and concluded that these early European hunter-gatherers were using a system of intentional and conventional signs on mobile artefacts.
This is directly relevant to one of the fundamental propositions of my research: that a portable prehistoric object can carry a socially understood visual system, with recurring signs or units deliberately incorporated into the object.
Their system is not my proposed system. The researchers analyse geometric signs statistically and do not identify the pictorial vocabulary I describe in worked stone. Nevertheless, by 2026 archaeological research is explicitly describing conventional sign systems carried on mobile Palaeolithic objects.
My hypothesis adds another possibility: that some recurring signs are not merely geometric units but recognisable figurative subjects, anatomical details, orientations and transformations.
Bentz & Dutkiewicz (2026) — Conventional Signs on Mobile Aurignacian Artefacts
The Global Claim: One Code, Many Times and Many Makers?
The worldwide dimension was already present in my early hypothesis. The archived 2013 site invokes the “House of Babel”, describes the same iconography recurring throughout the material, and proposes basic “words” occurring in stone beyond one local collection.
A central part of my hypothesis has always been that comparable visual conventions occur in worked stone from widely separated regions. If that recurrence can be demonstrated systematically, several explanations would have to be tested: deep cultural inheritance, transmission between populations, recurrent constraints of stone and perception, independent invention, or some combination of these.
Crucially, global recurrence does not require every example to be the same age or made by the same species. A long-lived convention could be inherited, reinvented or transformed across immense spans of prehistory. That possibility makes careful dating and specimen-level context more important, not less. The conventions appear to be global, thumbs, half creatures, genital placement, composite images built into ambiguous illusions.
A Modern Functional Analogue: Portable Object + Orientation + Learned Code
A useful ethnographic comparison comes from mobile hunter-gatherer communication systems in which portable objects carry conventional information. Among the Punan Batu of Borneo, message sticks can use direction, length and attached signs to communicate movement, distance, identity, food requirements, resources and warnings.
This does not prove that prehistoric stones were used in the same way. It demonstrates something narrower but important: a portable object whose orientation and conventional features are learned by a community can function as a practical communication device in a mobile hunter-gatherer context.
What Would Make the Hypothesis Stronger?
- Independent recurrence: the same proposed unit occurs on unrelated finds.
- Stable position: eyes remain eyes, ears remain ears, and geometric modifiers occupy comparable locations.
- Operational repeatability: rotation, flipping or pointing changes the reading in the same way on multiple objects.
- Technological integration: the imagery is structurally tied to deliberate reduction, edge production, engraving, pigment or tool design.
- Blind comparison: observers can identify recurring structures before receiving detailed prompting.
- Chronological discipline: each specimen is dated and contextualised independently rather than assigned a common age.
- Maker discipline: no species is assumed merely because a preferred evolutionary narrative expects it.
- Global comparison: putative matching conventions are tested against genuinely independent assemblages.
Conclusion: A Communication Ecology, Not One Universal Stone Dictionary
Taken as a whole, this is the communication hypothesis I have been developing and demonstrating since 2012. It has always included different kinds of objects performing different apparent functions. Some may have operated as immediate survival devices. Some may have selected animal identity or direction. Some may have animated movement. Others may encode conceptual, reproductive, social or narrative relationships. Some objects may have carried considerably more information than others, and some may ultimately prove not to have been communication devices at all.
Nor do these objects have to belong to one date, one profession, one population or even one hominin species. What connects the research is the recurrence of apparently intentional visual conventions capable of representing subjects, details, states, directions and relationships.
The recent Boxgrove research by Shipton and colleagues provides an important archaeological parallel. Their work identifies both a directly observable social vocabulary in standardized handaxe features and a separate route toward symbolic language through the teaching requirements of difficult stone technology. Experimental knapping research likewise shows why teaching and verbal/proto-verbal communication become increasingly useful as technological concepts become abstract and difficult to observe.
My hypothesis is that in some artefacts the informational system did not vanish completely with the voices of the makers. Part of it may still be there in the stone.
FAQ: Prehistoric Language, Communication and a Visual Lexicon in Stone
What is the central hypothesis?
I propose that some prehistoric worked stone and other portable artefacts incorporated recurring visual conventions that could function as a material communication system. Recognisable subjects, anatomical details, orientation, rotation, shared outlines and other repeated features may have acted as information-bearing units within a visual lexicon.
Is this a new hypothesis developed after the recent Boxgrove paper?
No. I was publicly describing a hidden language element, recurring glyphs and flint artwork language in 2012, while my own contemporaneous account places the beginning of those observations in 2011. The surviving 2013 Revelation in Stone archive preserves the communication-device hypothesis in substantially greater detail.
Does every proposed communication artefact work in the same way?
No. That has never been the proposal. Some objects appear comparatively simple and may identify little more than a subject, direction or animated action. Others appear to contain several selectable subjects, anatomical details, orientations or transformations. The hypothesis concerns a family of related visual conventions and device types rather than one rigid universal format.
Does every object have to carry a complicated message?
No. A communication system can contain both simple and complex units. One object might communicate little more than PERSON — WALKING or ANIMAL — THIS WAY, while another may combine subject, direction, gaze, feeding, alertness or other information.
Were all of these artefacts made at the same time?
No such assumption should be made. Individual objects may differ greatly in age, context and function. Apparently related conventions could potentially be separated by thousands, hundreds of thousands or much greater spans of time. Each assemblage has to be considered on its own archaeological evidence.
Were all of the makers Homo sapiens?
That cannot be assumed in deep prehistory. Depending on the age and context, different hominin populations may have been responsible. Similar visual conventions do not prove one species, one population or one continuous culture.
Does recurrence imply that the same culture made everything?
No. Recurrence is something to investigate, not proof of direct cultural continuity. Similar conventions could result from long-lived transmission, interaction between populations, inheritance, independent reinvention or other processes. Dating and context have to be established separately from visual comparison.
What do I mean by a visual lexicon?
I mean a restricted repertoire of recurring visual units that appear to represent recognisable concepts. My proposed vocabulary includes animals, humans, body parts, geometric forms, directional features and behavioural modifiers. If the same units recur intentionally and conventionally, they become candidates for lexical components rather than isolated pictures.
Does an image of an animal prove there was a spoken word for that animal?
No. A conventional image does not preserve the sound that prehistoric people used for it. But if a population intentionally distinguished animals, body parts, actions, directions and behavioural states through a shared visual vocabulary, that would be consistent with a corresponding verbal or proto-verbal conceptual lexicon.
Why do I think spoken language may have existed alongside the visual system?
A visual code first has to be learned. Speech and gesture provide efficient ways to teach which subject or detail matters, what an orientation means and how another person should interpret the object. Experimental stone-tool research likewise shows that active teaching, including verbal teaching, can greatly assist the transmission of difficult technological knowledge.
If spoken language existed, why would anyone need a stone communication device?
Because the two systems could serve different purposes. Speech could teach and discuss the code, while a portable visual object could preserve information, demonstrate it, select between alternatives or communicate without speaking when silence was useful. The material device therefore supplements language rather than replacing it.
What does “Language teaches the code; the stone carries the code” mean?
It means that people could use speech, gesture and demonstration to teach shared visual conventions. Once those conventions were understood, the artefact itself could carry them. Pointing, rotation, flipping or orientation could then select particular information, potentially without requiring speech at the moment of use.
What is the “dinner and danger” idea?
It is a simple way of describing the survival value of information about animals. An animal may represent food, danger, competition or opportunity. A portable visual referent could potentially identify what animal was present, where it was, what direction it was travelling and whether it appeared alert, feeding or resting.
Were all of the proposed devices hunting aids?
No. Hunting and survival provide an obvious practical use for some objects, particularly where silence would matter, but other artefacts may have been used for teaching, social identity, storytelling, memory, ritual, classification, amusement or functions that are no longer recoverable.
How could rotating or flipping a stone communicate information?
In some objects I identify multiple subjects or profiles arranged so that physical manipulation changes what is presented. Rotating a stone can select a particular profile or orientation; flipping it may reverse the direction in which a subject appears to face. If those relationships were intentional and shared, manipulation itself becomes part of the communication system.
What do Shipton and colleagues actually argue at Boxgrove?
Their argument has two important elements. First, standardized features such as the tranchet tip were directly observable social norms: culturally shared features that members of the Boxgrove group recognised and reproduced. Second, the difficult technique of turning-the-edge is visually and causally obscure enough that teaching it may have required symbolic communication and conventional referents. Both routes lead the authors toward symbolic language.
Is the tranchet tip itself part of the Boxgrove social vocabulary?
Yes, in the sense used by Shipton and colleagues. The standardized tranchet feature was directly visible on the handaxe and formed part of a culturally understood technological norm. This is distinct from the turning-the-edge argument, where symbolic language is inferred from the difficulty of transmitting an underlying technique.
Did Shipton and colleagues discover actual prehistoric words?
No. They do not recover the sounds or exact gestures used at Boxgrove. Their research identifies socially standardized tool features and argues that difficult technological concepts probably required conventional symbolic referents. The prehistoric forms of those referents have not survived.
Does the Shipton et al. Boxgrove research support verbal language?
It supports symbolic language more broadly and is consistent with verbal or proto-verbal communication. Turning-the-edge may have required specific conventional referents because it is difficult to teach through observation or pointing alone. Spoken language is therefore a strong possibility, although archaeology cannot recover the actual prehistoric speech.
How is my proposed lexicon different from the Shipton et al. Boxgrove vocabulary?
Their work identifies culturally standardized technological features and infers symbolic concepts required for teaching. My hypothesis proposes an additional pictorial layer: recognisable subjects such as animals, humans and body parts, together with orientation, anatomical details and transformations that may themselves have carried information.
So is the Boxgrove vocabulary visible or inferred?
Both elements are present in the Shipton et al. interpretation. The tranchet tip is a directly observable standardized social feature. The symbolic referents required to explain and teach difficult technological concepts such as turning-the-edge are inferred. This distinction is important.
What is my proposed “egg” word at Boxgrove?
I argue separately that the recurring ovate handaxe form closely reproduces the geometry of an egg, particularly a goose egg, and that this relationship can be examined geometrically rather than relying only on visual resemblance. Within my proposed pictorial lexicon, EGG is therefore one of the clearest candidate visual concepts. This is my interpretation, not a conclusion of Shipton and colleagues.
Do the Boxgrove researchers identify elephants, apes, hands, eggs or my other pictorial motifs?
No. Those identifications are mine. I recognise such recurring visual units within Boxgrove and other worked-stone assemblages, but they should not be attributed to the Shipton et al. paper unless the authors themselves explicitly make such an identification.
Why are the Bizmoune shell ornaments relevant?
The Bizmoune objects provide another example of deliberately selected and modified portable objects interpreted archaeologically as socially meaningful symbolic signals. Their exact communicated content is unknown. I additionally recognise within some published examples recurring visual forms that resemble elements of my previously proposed vocabulary.
Do the Bizmoune researchers identify my swan, elephant, thumb or other motifs?
No. Those visual readings are my interpretation. The archaeological research establishes the deliberate selection, modification and symbolic/social signalling function of the shell ornaments. My separate question is whether some of their surviving visual structure may preserve part of the information that was being communicated.
Why is a restricted visual repertoire important?
If almost any object could be interpreted as almost any image, the lexicon hypothesis would be weak. The stronger prediction is recurrence of a comparatively restricted set of subjects and modifiers, appearing in comparable locations and sometimes in repeated combinations. A limited vocabulary is therefore more significant than an unlimited number of possible pictures.
Could natural stone or shell shapes contribute to the imagery?
Yes. My hypothesis does not require every line or curve to have been created from scratch. A maker could recognise and deliberately exploit a useful natural form, then modify, sharpen, engrave, pigment or combine selected features to complete or transform an image. The archaeological question is whether deliberate working is systematically integrated with the proposed visual structure.
How can deliberate imagery be distinguished from accidental resemblance?
A single resemblance is weak evidence. The case becomes stronger when optional details recur in anatomically appropriate positions, when the same subjects and combinations appear across independent objects, when working marks contribute directly to those forms, and when rotation or orientation repeatedly produces comparable visual operations.
What evidence would strengthen the hypothesis?
Repeated motifs on independently recovered artefacts, stable placement of visual units, demonstrable technological integration, repeatable orientation or rotation effects, microscopic evidence of deliberate modification, pigment analysis, blind recognition studies and well-dated archaeological contexts would all strengthen the case.
What evidence would weaken the hypothesis?
The hypothesis would be weakened if proposed motifs failed to recur under controlled comparison, if observers could generate equally persuasive unlimited alternative images, if supposedly meaningful details proved unrelated to deliberate working, or if claimed repeated conventions disappeared when tested against independent assemblages.
Is pareidolia enough to dismiss the proposed images?
No psychological label by itself determines how an artefact was made. Pareidolia is a possible alternative explanation for perceived imagery, but it has to be tested against recurrence, deliberate modification, technological structure, placement and repeated visual relationships. The issue is whether the pattern survives objective comparison.
Is the proposed visual communication system global?
My hypothesis is that comparable visual conventions occur internationally. That was already part of my early research. However, international recurrence does not mean every object is the same age, belongs to the same culture or was made by the same hominin population. Those questions must be tested independently.
Would a worldwide recurring vocabulary prove one universal prehistoric language?
No. A recurring visual convention could potentially survive across different spoken languages, populations and periods, just as some modern visual signs can cross linguistic boundaries. Demonstrating recurrence would raise questions about transmission and cultural longevity; it would not by itself demonstrate one universal spoken language.
Why call these objects communication devices rather than simply art?
Art and communication are not mutually exclusive. My use of “communication device” refers specifically to objects where imagery appears structured so that subjects, orientations or details could be selected, demonstrated or manipulated to convey information to another person. Some objects may have had artistic, practical and communicative roles simultaneously.
Does this hypothesis require prehistoric people to think like modern humans?
No. It requires only that makers could learn shared conventions, recognise subjects and relationships, intentionally modify material and understand how other members of their group would interpret those conventions. The precise cognitive and linguistic capacities involved may have varied greatly through time and between hominin populations.
What is the most important claim being made on this page?
Not that every proposed image has already been proven or decoded. The central claim is that recurring worked forms may preserve evidence of a structured material communication system, and that some of the information carried by that system may remain physically visible in the artefacts rather than surviving only as an inferred, vanished language.