Phylogeny as the basis for the sub-fields of biosemiotics

1 Introduction: biosemiotics and biology

The study of communication is an important part of semiotics (Favareau 2010; Kull 1999, 2022). Since communication is a fundamental feature of life, Anderson et al. (1984) argued that the communication processes of all life should be included within biosemiotics. Sebeok (2001) maintained that semiosis and life are coextensive “because there can be no semiosis without interpretability.” Biosemiosis and life are coextensive because all forms of communication – the production of signs and the interpretation of their meanings – are based on different physiologies that have evolved along with genes and species in the tree of life. Biosemiotics sees the molecular evolution of life and the evolution of semiotic systems as two aspects of the same process (Queiroz et al. 2011). These ideas have become a template for contemporary biosemiotics (Anderson 2016: 278–279; Cobley et al. 2017).

Here I develop these ideas and argue that the sub-fields of biosemiotics can be based on the phylogenetic branches of the tree of life (Figure 1). I call this connection between biosemiotics and the phylogenetic tree of life, the “phylosemiotic hypothesis.” According to this hypothesis, phylogeny is the scaffolding that should be used to support the sub-fields of biosemiotics. In a phylogenetic tree such as Figure 1, genetic distances represent how different two lineages have become. Some genes diverge more rapidly than others and therefore molecular clocks do not all tick at the same rate, but there are useful ways to correct for these differences (Kapli et al. 2020). Genetic distances represent differences in biochemistry, physiology, and sensory mechanisms that send and receive signals. The phylosemiotic hypothesis is that these genetic distances can be reasonably interpreted as biosemiotic distances or how biosemiotically different these species are. This hypothesis offers a Darwinian approach to understanding the categories of biological communication. Potential problems with this hypothesis are discussed in Section 5.

Figure 1:

The phylogenetic tree of life modeled as a hierarchical tree of divergences. The phylogenetic distances between species in this tree are based on differences in DNA sequence (Kapli et al. 2020; Li 1981). The highest-level divergence is on the left between Archaea and Bacteria. The various colors reflect some of the lower-level divergences. Eukarya (eukaryotes) are a branch of Archaea. Further down in the hierarchy, Eukarya are divided into three groups: yellow (Diaphoretickes, including plants), green (Amorphea which includes Fungi, Metazoa and Amoebozoa), and purple (Euglenozoa). The dark purple arrow points to the divergence of plants and animals about 1.6 billion years ago (Lineweaver and Chopra 2019). The labels “zoosemiotics” and “phytosemiotics” are associated with branches where most biosemiotic efforts have been focused. However, these biosemiotic efforts have not used such detailed phylogenetic trees to specify what life forms are under discussion. The realm of zoosemiotics could be the biosemiotics of animals circumscribed in red. The realm of phytosemiotics could refer to the biosemiotics of a subset of Diaphoretickes known as Archaeplastida (circumscribed in blue). However, reducing the biosemiotics of Eukaryotes to only phytosemiotics and zoosemiotics excludes most eukaryotes and therefore excludes most of the diversity of eukaryotic communication. This figure is a modified version on Figure 5 of Lasek-Nesselquist and Gogarten (2013) based on ribosomes.

2 How the sub-fields of biosemiotics map onto the tree of life

There is no consensus about how the sub-fields of biosemiotics should be organized. In the literature one can find mention of phytosemiotics and zoosemiotics (Sebeok 1968). A naïve interpretation of these terms is that they are mutually exclusive sub-fields circumscribed by the blue and red lines in Figure 1. Alternatively, vegetative semiosis can refer to the lowest level of a three-level nested hierarchy of vegetative, animal and cultural semiosis, in which cultural semiosis is the highest level and includes the capabilities of lower levels, just as zoosemiosis includes the semiosis of the lower vegetative level (Kull 2009, 2014). This three-level nested-hierarchy can be compared with the two-levels of semiosis of Sharov and Vehkavaara (2015): (1) a “primitive kind of semiosis,” which they call “protosemiosis” and (2) an “advanced kind of semiosis,” which they call “eusemiosis.” Assuming a similar evolutionary progress in semiosis from primitive to advanced, Hoffmeyer and Stjernfelt (2016) describe an eleven step “great chain of semiosis”: (1) molecular recognition, (2) prokaryote-eukaryote transformation (privatization of the genome), (3) division of labor in multicellular organisms (endosemiosis), (4) from irritability to phenotypic plasticity, (5) sense perception, (6) behavioral choice, (7) active information gathering, (8) collaboration, deception, (9) learning and social intelligence, (10) sentience, (11) consciousness. More thematically, without explicitly invoking semiotic progress, Queiroz et al. (2011) describe six branches of semiotics: (1) zoosemiotics, (2) cellular and molecular semiotics, (3) phytosemiotics, (4) endosemiotics, (5) semiotics of neurobiology, and (6) origins of semiosis and semiotic thresholds. Branches (4), (5), and (6) of Queiroz et al. (2011), as well as ecosemiotics do not fall within the scope of this article and will therefore not be discussed.

Authors using the term “phytosemiotics” have paid little attention to phylogenetic trees such as Figure 1. For example, Colapietro (1993: 160) has defined phytosemiotics as “the semiotics of plants.” The working definition of phytosemiotics in Kull (2000: 328) is multicellular organisms with cellulose walls (i.e., bryophytes and vascular plants but not algae). More broadly, Sebeok (1997) has argued for using the five kingdoms of life – a pre-phylogenetic tree taxonomy proposed by Whittaker (1969).

Biosemiotic work done on bacteria can be classified using the major branches of bacteria. For example, in Figure 1, bacteria have been divided into two main groups labeled “clade 1” and “clade 2 + clade C” (colored light purple and aqua respectively). Biosemiotic work on Archaea can be similarly classified using the major branches of Archaea. Biosemiotic work on protists can be classified using the branches of Eukaryotes shown in Figure 4. Although viral phylogeny is a work in progress and has not been included in the tree of cellular life (Figure 1), biosemiotic work on viruses can similarly be classified using the major branches of viral phylogeny. See for example Figure 7 of Nasir and Caetano-Anollés (2015). Such biosemiotic work would fall under Queiroz et al.’s branch (2): cellular and molecular semiotics.

Life on Earth has been evolving and diverging for 4 billion years. Plants and animals diverged about 1.6 billion years ago. The common ancestor of plants and animals lived about 1.6 billion years ago. The modern phylogenetic tree of life contains dozens of independent branches that diverged earlier than that. Thus, among the life forms on Earth, we can expect dozens of differences in signaling as important as the signaling differences between plants and animals.

Fungi and Amoebozoa have communication systems that are hundreds of millions of years more closely related to animals than they are to plants (see Figure 1). If the differences between plants and animals and their abilities to make and receive signals is large enough to recognize separate levels of semiotics (i.e., zoosemiotics and phytosemiotics) then there are many dozens of other pairs of independent branches that diverged from each other earlier than plants and animals did. It is reasonable to expect that the physiologies, signaling, receptors and gene regulators of these pairs are as different from each other as those of plants and animals are (Chang and Stewart 1998; Jung 2007). The phylosemiotic hypothesis suggests that if one is trying to identify the most characteristic differences in the modes of communication, one should use the branches of the tree of life. This would create a hierarchy of sub-divisions of biosemiotics with many dozens of divergences as important as the difference between the semiotics of animals and the semiotics of plants (Figure 2).

Figure 2:

This diverging phylogenetic tree is a simple version of the tree of life in Figure 1. On the left, plants and animals diverged from each other 1.6 billion years ago (blue dot). At the top, three other analogous pairs of organisms are indicated by the red and green bars. The pairs are labeled “1,” “2,” and “3.” The red and green bars represent sister clades: two groups of organisms that diverged from each other at the purple dots between 1.6 and 2 billion years ago. Thus, for each pair, the red organisms diverged from the green organisms earlier than animals and plants diverged from each other. The common ancestors of each of the red and green pairs is marked with a purple dot that pre-dates the common ancestor of animals and plants (blue dot at 1.6 billion years ago). The result of more than 1.6 billion years of separate evolution means that comparing animal communication to plant communication is easier than comparing the communication of the organisms in red to the communication of the organisms in green. Horizontal gene transfer complicates this divergence-only tree.

If one is interested in the differences in the production and interpretation of the signals of two organisms, one needs to look at their evolutionary paths. What biochemistry or sense organ is making the signal? What biochemistry or sense organ is receiving the signal? Biosemiotics is the study of biosemiosis and biosemiosis is based on genes, enzymes, hormones, receptors, regulators and sense organs – all of which have evolved and diverged independently in many dozens of branches over the past 1.6 billion years (Figures 1–3).

Figure 3:

A more recent and more comprehensive tree of life than Figure 1. Like Figure 1, it shows the two main domains of life (Bacteria and Archaea) with Eukaryotes branching within the Archaea. The dashed blue line is a crude estimate of the time 1.6 billion years ago when plants and animals diverged from each other. Communication differences as profound as the differences between animals and plants should be present in any two lineages that diverge within the dashed blue line. Such clades are today more phylogenetically distant (and arguably more biosemiotically distant) from each other than plants and animals. This figure is a modification of Figure 1 of Hug et al. 2016.

Taking the phylosemiotic hypothesis (branches of biosemiotics should roughly mirror phylogenetic branches) seriously, would require biosemioticians to become microbiologists familiar with the diversity of the biochemistry of communication at the cellular and molecular level in the hierarchical branching of the tree of life. Much work has already been done in what Queiroz et al. (2011) call branch 2 of biosemiotics. This work includes quorum sensing in bacteria (Brambila-Tapia and Pérez-Rueda 2014; Waters and Bassler 2005; Wei et al. 2020), signal transduction in Bacteria, Archaea, and single-celled organisms (Matilla et al. 2022; Suzuki et al. 2010; Zhong et al. 2017), and comparative protein phylogeny (Capra and Laub 2012; Zscheidrich et al. 2016).

The phylosemiotic hypothesis predicts that the biggest difference in communication would be found between the earliest diverging branches. This is an important general idea that needs to be tested. For example, it predicts that the difference in communication between a bacterium and an archaean is greater than the difference in communication between two bacteria or between two Archaea. Verifying this will require much detailed microbiology and some agreement on how to quantify chemical differences in communication. The phylosemiotic hypothesis also predicts that the difference in communication between a bacterium and an archaean is greater than the difference between two eukaryotes: Amoebas and Apes for example. I suggest that this may be true because of the fundamental molecular differences between Bacteria and Archaea, and the fundamental molecular similarities of eukaryotes. But to verify this, we will need some agreed-upon metric in the space of communication differences that would give appropriate weightings to differences at the levels of molecules, proteins, cells, organs and the whole organism. To date, I know of no such consensus.

Just as a coherent phylogeny is predominantly composed of monophyletic (not paraphyletic) groups, a coherent biosemiotics should be based on the same monophyletic groups. For example, at the highest cladistic level, biosemiotics would be more consistent with the basic details of cellular communication if it were divided according to Figure 1 into archeosemiotics and bacteriosemiotics (the semiotics of Archaea of Bacteria, respectively). As one simple example, the cellular communication necessary to form biofilms is different between Archaea and Bacteria (van Wolferen et al. 2018), but biofilms made of both Archaea and Bacteria do exist (Fröls 2013). At lower cladistic levels, in Figure 1, Eukarya are divided into Diaphoretickes (yellow), Amorphea (green), and Euglenozoa (purple). This suggests that semiotics of eukaryotes includes sub-categories of communication specific to Diaphoretickes, Amorphea, and Euglenozoa, respectively. These divisions would not only expand the horizons of biosemiotics but also increase its consistency with the diversity of biocommunication. Both reptiles and mammals, for example, hear sounds, but during their 310 million years of independent evolution, the architecture of the ears (and presumably the sound processing in the brain) have diverged (Smith and Takasaka 1971).

The number of divergence points within the grey dashed circle of Figure 4 are within the timeframe ∼1.6 > t > ∼1 Gya (Lineweaver and Chopra 2019, but see Chernikova et al. 2011; Strassert et al. 2021). These divergence points are analogous to the divergence points marked in purple in Figure 2 that have a timeframe 2 > t > 1.6 Gya. To enumerate the independent lineages that cross the grey dashed circle (and therefore have been independent for more than a billion years) we count 6 lineages of Amoebozoa, 1 lineage of Animals (shown in Figure 5), 3 lineages of Fungi, 7 lineages of Excavates and 27 lineages of Diaphoretickes. These 44 lineages of Eukaryotes have signaling apparatus that has diverged from each other for more than a billion years. Eukaryosemiotics is an enormous field of study that will benefit by specifying the large differences in the semiosis between each of these lineages.

Figure 4:

The tree on the left is a reduced version of Figure 3. The circular tree on the right is a more detailed look at divergences within Eukaryotes. The purple arrow near the center points to the divergence of plants and animals about 1.6 billion years ago (Lineweaver and Chopra 2019). Here we have a typical example of what is meant by a “hierarchy of phylogenetic branches.” The three largest clades of Eukaryotes indicated by the arcs in the outer circle are Diaphoretickes, Excavates, and Amorphea. Within the Amorphea are the Amoebozoa (yellow) and Opisthokonta (blue). Within the Opisthokonta are Fungi (small magenta font) and Animals (small red font). Within the animals, four clades are shown: Porifera, Bilateria, Coelenterata and Placazoa. The smallest circle (grey dashed) indicates the time roughly about 1 Gya and corresponds to the origin of animals. Figure 5 shows more details of the phylogenetic branch (“Animals” and “Bilateria” above) that led to humans. The center of the circle corresponds to the origin of Eukaryotes about 1.6 billion years ago (Lineweaver and Chopra 2019). The circular figure on the right is a modified version of Figure 1 of Burki 2014 (see also Burki et al. 2020).

Figure 5:

Approximately one billion years of animal phylogeny. This is a more detailed look at the lower right portion of Figure 4 labeled “Animals.” If zoosemiotics is about understanding the most important details of animal communication, then the divisions within zoosemiotics should mirror the branches of the phylogeny of animals. This animal phylogeny underestimates the diversity of animal communication because in it, once a group branches off from the lineage that led to us, no further branching is shown. For example, our common ancestor with Cnidaria (corals and jellyfish) lived about 800 million years ago. This is when the ancestor of the 9,000 extant species of radially symmetric Cnidarians diverged from the ancestor of the bilaterally symmetric animals, Bilateria. Fifty million years later the Bilateria diverged into deuterostomes (e.g., humans) and protostomes (e.g., flies). The diversity of some of the Bilateria that led to humans is shown. However, all 9,000 species of Cnidarians are represented by one diagonal line. The next diagonal line represents more than a million protostome species (Dawkins and Wong 2016). The biosemiosis in our own lineage has depended on the evolution of heads, mouths, ears, eyeballs, and hands.

3 Anthroposemiosis

Many semioticians were only interested in studying human communications without worrying about other species. For them, the study of human communication (aka anthroposemiosis or cultural semiosis) was the only kind of semiosis. Kull (2009) proposed three divisions: vegetative, animal, and cultural semiosis. Treating human communication separately and higher than the communication of other animals is standard biosemiotic practice and has been justified based on a scale of “semiotic thresholds” (Cobley et al. 2017; Deely 2005, 2010; Nöth 2000; Sebeok 1997).

Kull (2009) has argued that there is a “lower semiotic threshold [that occurs] at the appearance of a living cell with its agency, memory, inside-outside and self-other distinctions…” Hoffmeyer and Stjernfelt (2016) parse semiotic evolution into an eleven step “great chain of semiosis” from primitive to advanced levels with humans being the most advanced. These hierarchical structures with higher and lower levels can be contrasted with Darwin’s (1837) view: “It is absurd to talk of one animal being higher than another.” See also Gould’s critique of the general idea of evolution as progress from lower to higher (Gould 1981, 1989, 2002).

Language is often considered as completely unique to humans. Linguists Hockett (1960) and Pinker (1994: 332–369) emphasize the large differences between human and non-human animal communication. Sebeok has proposed to not use the word “language” to describe animal sign systems (Martinelli 2010). Ben-Yami (2017) says humans are the only species with language.

Other linguists and biologists articulate a more nuanced assessment in which human communication is taken to be the most advanced in a continuum of overlapping animal communication systems – an approach that aligns with gradualist evolutionary principles (Amphaeris et al. 2022). Hauser et al. (2002) discuss the evolution of systems that enable different kinds of languages and support the idea of an evolutionary continuum between the communication methods of non-human animals and humans. Some scholars believe that we don’t yet know enough about non-human animal communication to claim that they don’t use or understand “language” (Ten Cate 2017).

An anthropocentric approach to biosemiotics is to use human language as the definition of language. After assuming this human standard, the question is asked: Which species communicate in a way that is most similar to human language? The phylosemiotic prediction is that species closest to us in the tree of life will have the most human-like communication. Non-human communication scholars have focused on chimps, gorillas, orangutans, dolphins, orcas, and corvids. All of these are closely related to humans since they are found in the tiny branch of eukaryotic bilateria known as tetrapods. They have common ancestors with humans 7, 10, 16, 90, 90, and 310 Mya, respectively (Figure 5).

With a less anthropocentric approach, the phylogenetic tree can be used to parse the broad patterns in the biological hierarchy of communication systems. The common ancestor of animals and fungi lived about 1.2 billion years ago, while the common ancestor of animals/fungi and plants lived about 1.6 billion years ago (Lineweaver and Chopra 2019). Thus, after diverging from plants, the genetic, molecular, and physiological basis for animal/fungi biosemiosis evolved identically for about 400 million years. This 400-million-years of common evolution inevitably produced some new biosemiotic/molecular features shared by animals and fungi (= Opisthokonts) but not by plants. These include extracellular digestion and the ability to absorb nutrients by osmosis (heterotrophy) and the ability to synthesize extracellular chitin in cell, cyst or spore walls. The name “Opisthokonts” means having a single posterior flagellum. The resulting direction of locomotion would constrain communication, just as the anterior heads of vertebrates do. Thus, the phylosemiotic hypothesis predicts that fungi communication is more similar to animal communication than it is to plant communication. And similarly, that Amoebozoa communication is more similar to Opisthokonta communication than it is to Diaphoretickes communication (Figure 4). These potential phylogeny-dependent patterns of biosemiosis have yet to be investigated with the exception of some biosemiotic analyses of plant and fungal communication (Comollo 2024; Hiernaux 2019; Witzany 2008). Notice, however, that this modest approach does not depend on any normalization to human language and is consistent with Dobzhansky’s (1973) adage: Nothing in biology makes sense except in the light of evolution. This more modest approach is to be preferred by any science with aspirations towards descriptive objectivity. It encourages investigation of other species and opens up biosemiotics to an immense diversity of communication systems.

For centuries we defined ourselves as “man the tool user.” But when Louis Leakey heard about Jane Goodall’s 1960 discovery of tool use among chimpanzees, he wrote “Now we must redefine tool, redefine man, or accept chimpanzees as human.” Similarly, for centuries, without having thoroughly studied language use in other species, we have defined ourselves as “man the language user” (Ben Yami 2017; Pinker 1994: 332–369). However, as the communication of more species is looked at more carefully, this notion (just like the notion of our unique use of tools) is increasingly being questioned (Bakker 2022; Cholewiak et al. 2013; Doyle 2006; Doyle 2009; Doyle et al. 2011; Doyle and Sharpe 2021; Ferrer-i-Cancho et al. 2022; Hauser 2000; Wohlleben et al. 2016).^[1] To paraphrase Leakey: Now we must redefine language, redefine man, or accept dolphins, orcas, corvids, bees, and many species of trees as human (see however Pinker 1994).

4 Bat and whale echolocation … and cortical mapping

In this section we describe two examples of how phylogenetic branches can be used to identify important biosemiotic and functional differences. Bats (Chiroptera) are known for their ability to echolocate. The deepest phylogenetic division in the bats is between the megabats and microbats. Our phylosemiotic hypothesis predicts that there would be an important difference between the ways these two groups communicate. There is. Microbats use laryngeal echolocation to emit high-frequency signals to detect prey and avoid obstacles (Teeling et al. 2002). No megabats use laryngeal echolocation. However, one genus of megabat (Rousettus) produces signals in a very different way; by clicks of the tongue (Herbert 1985; Holland et al. 2004; Möhres and Kulzer 1956). With a precise motor control of their tongues, they are able to aim sonar beams in different directions without moving their heads (Halley et al. 2022). The other genera of megabats navigate solely by vision and smell during flights to and from foraging grounds at dusk and dawn. All megabat eyes have high densities of rod photoreceptors; an adaptation for nocturnal vision. Bat phylogeny and the differences in communication between bat clades is still a work in progress (Nojiri et al. 2021; Veselka et al. 2010; Wittrock 2010).

The deepest division in the Cetacea is between baleen whales (Mysticeti) and toothed whales (Odontoceti). Toothed whales include porpoises, dolphins, beluga whales, sperm whales, beaked whales, and other predatory whales. Baleen whales are filter feeders and include blue whales, humpback whales, and bowhead whales. Analogous to our phylosemiotic prediction in bats, there should be an important difference between the ways toothed and baleen whales communicate. There is. Only toothed whales have echolocation capabilities. They use nasal air sacs (melons). For example, dolphins make a broad range of sounds: frequency-modulated whistles, burst-pulsed sounds, and clicks. The clicks are directional and are used for echolocation. The other sounds are for communication. Baleen whales on the other hand make vocalizations (i.e., they sing) but do not echolocate. Toothed whales echolocate and communicate at high frequencies while baleen whales sing at low frequencies (e.g., the songs of the humpback whale, Doyle and Sharpe 2021). Toothed whales use ultrasonic frequencies 20 kHz to few GHz; baleen whales use infrasonic frequencies 0.1 Hz–20 Hz (humans communicate at “normal” sound frequencies in between these extremes: 20 Hz to 20 kHz Viglino et al. 2021).

Biosemiotics is concerned with how signals are made and how they are interpreted by the brain (if the organism has a brain). The surface area of the cortex assigned to signal processing is a good proxy of the significance that the organism assigns to different means of communication (Figure 6). The identification of a wide variety of tetrapod cortical mapping would be a constructive place to start tetrapodsemiotics. Imagine different distributions for other encephalated species in which vision or olfaction play no role (Niimura et al. 2014) or in which, as in snakes, chemoreception, taste, and heat sensitive organs add to vision. In echolocating bats, behaviorally important tones have a disproportionately large cortical representation (Neuweiler et al. 1980; Suga and Jen 1976).

Figure 6:

Two illustrations of how resources in the mammalian cerebral cortex are distributed to different parts of the body. The top left (a) illustrates sensory nerves or inputs into the human brain, while the top right (b) illustrates motor skills (Penfield and Rasmussen 1950: 44, 57). Notice the large region in (b) dedicated to “vocalization” as well as to the manual dexterity (arguably, this distribution makes sign language the default for humans who can’t speak). The bottom illustration (c) represents the distribution of the somatosensory cortex of a star-nosed mole to different body parts. Star-nosed moles are nearly blind mammals that live underground in the dark. Tactile sensory organs are numerous in each of the rays around the nose. Their innervation density is higher than any other mammalian skin surface (Figure 9 of Catania and Kaas 1997).

5 Complications and arguments against the phylosemiotic hypothesis

If we want to base biosemiotics on the phylogeny of life it is important to recognize that the tree of life is more complicated than just a tree of divergences (e.g., Figure 2). There is much horizontal gene transfer (HGT; Doolittle 2000; Doolittle and Brunet 2016). Endosymbiosis plays the leading role in the origin of eukaryotic cells (Sagan 1967). These mechanisms undermine the diverging tree metaphor and make a reticulated network a more appropriate metaphor since a reticulated network can include converging branches to represent endosymbiosis (Gontier 2015, 2020). The phylosemiotic hypothesis is not restricted to a divergence-only tree of life. Rather, the detailed complications of the phylogeny provide a detailed complicated scaffold for the sub-fields of biosemiotics.

What about inter-lineage semiosis and exosymbionts? The evolution of endosymbionts (alpha-proteobacteria becoming mitochondria) required the evolution and co-evolution of communication between different clades (presumably Bacteria and Archaea). In this case, the method of communication between the clades can be seen as a unit of selection (cf. Okasha 2006). The communication between our bodies and our gut microbes must also have been the result of co-evolution. A sheep dog can communicate to a flock of sheep. This communication, like other inter-lineage semiosis, must have co-evolved as a unit of selection from the time sheep and dogs came in contact with each other. See Musser et al. (2014) on cross-species vocal communication between orcas and dolphins. The phylosemiotic hypothesis suggests that the co-evolution of these semiotic abilities will have genetic correlates in both lineages.

In Figures 1–3, we see that some Asgard Archaea are more closely related to Eukaryotes than they are to other Archaea (e.g., Lineweaver 2020). Therefore, Eukaryotes are a branch of Archaea. However, Eukaryotes are a chimera of both Archaea and Bacteria. For example, eukaryotic reproduction seems to have a more archaean origin while eukaryotic metabolism has a more bacterial origin (see for example Figure 1 in Weiss et al. 2016). The fundamental differences between these deepest divisions in the tree of life include fundamental transitions in biochemistry and modes of communication. These transitions should be identified and their independent signaling abilities used to classify the branches of biosemiotics. The phylosemiotic prediction is that these transitions in communication are best navigated with a phylogenetic tree.

Species trees are not the only trees upon which biosemiotics can be scaffolded. Trees can be made from individual genes or from gene cassettes. Due to HGT and other kinds of gene mobility, a species tree and a tree of a particular gene are not necessarily identical. See, for example, Figure 1 of Puigbo et al. (2013). When particular genes are strongly associated with a means of communication (or interpreting signals) then biosemiotics can be scaffolded onto trees made from the lineages of these genes.

Since hippos are the closest living relatives of whales, a phylosemiotic prediction would be that hippo communication would resemble whale communication more than any other animal communication would (Maust-Mohl et al. 2018). One test of this is to compare the machinery used to make and interpret sounds in these two sister clads. Such tests, to the extent they have been done, support the phylosemiotic hypothesis (as described above for bats and whales).

The phylosemiotic hypothesis relies on genetic divergence being a quantifiable and reasonable proxy for the more difficult-to-quantify semiotic divergence. But toothed whales and microbats began to diverge genetically 80 million years ago and yet they both echolocate. Doesn’t that make their semiotic distance smaller than their phylogenetic distance? Haven’t their means of communication converged while their genetics diverged? Such cases of convergence (e.g., Morris 2003), if taken at face value, undermine the phylosemiotic hypothesis and the assumed correlation between phylogenetic distance and semiotic distance. However, if one looks more closely at the physiology that enables the echolocation, the evolution of the organs that produce the sound, the organs that receive the echoes, the frequency of the sound and the evolution of the neural circuitry that processes the echoes (e.g., Figure 6) one could find that these details and their detailed semiotic consequences have diverged with the genes and produced two different kinds of echolocation – with the exception of fundamentally conserved features – features of the common ancestor (deep homology) that enable the echolocation. Such cases would not be convergence but deep homology (e.g., Shubin et al. 2009).

How can one quantify communication differences? If one is narrowly interested in hydrodynamics, there is a smaller hydrodynamic distance between the shapes of dolphins and sharks, than between the shapes of dolphins and hippos. If one is narrowly interested in aerodynamics, there is a smaller aerodynamic distance between a maple seed and a helicopter. Thus, “distance” depends on what you are interested in. The phylosemiotic hypothesis suggests that if we are interested in the variety of communication, our focus should be on the evolution of the organs used to communicate and interpret signals. This would connect biosemiotics more coherently with evolutionary biology. The phylosemiotic hypothesis places more emphasis on the medium of communication and less on an abstract disembodied message. The phylosemiotic hypothesis suggests that higher level meanings are intrinsically embedded (and therefore cannot be removed from) the lower levels of semiosis. This view is similar to the view popularized by McLuhan in which the contents of a message are inextricably linked to the medium of the message: the medium is the message (McLuhan 1964, 1967).

One could argue that biosemiotics is a description of communication, not phylogeny. And that phylogeny is not a useful way to classify modes of bio-communication. Shouldn’t biosemioticians be more interested in the information theoretic analysis of the signals and what they communicate, rather than the detailed structures and particular frequencies responsible for the signals?

The phylosemiotic hypothesis is similar to what Kull (2009) has called the “taxonomic classification of the types of semiosis.” He doubts that such an approach can be used to classify functionality because the main power of semiotic modeling concerns the differences in the logical functions of semiosis, such taxonomical classifications based on historical (phylogenetic) divergences may not result in any profound functional typology (Kull 2009: 14; see also Jablonka and Ginsburg 2022). In contrast, I am arguing that phylogenetic divergences (and their correlated physiological differences) do lead to an important functional and physiological typology that provides a phylogenetic roadmap to navigate the broad expanse of non-human communication.

If social complexity leads to vocal complexity, then the important variable of complexity for a communication system will be correlated with the evolution of social complexity (i.e., in humans and eusocial insects (ants, termites)). If such complexity is not the result of a shared evolutionary history, then phylogeny will not be the best way to identify the similar styles of complex information exchange (Wilson and Hölldobler 2005). But if one looks more carefully at the details of the complexity of the communication, the phylosemiotic prediction is that phylogeny will become the most useful tool to characterize the communication. The identification of broad superficial functional similarities between species will not lead to a biosemiotics consistent with evolutionary biology. However, when one is willing to deeply parse the functionalities and pay attention to the details of how the organism sees or flies or echolocates, then fundamental, predominantly monophyletic groups begin to emerge from functional studies.

One might worry that adopting phylogeny as a scaffolding for biosemiotics would lead to too many empty shell sub-fields with little research activity. For example, Krampen (1992: 213) argued that if the phytosemiotic sluice was opened there would be no end to new semiotics – e.g., mycosemiotics, cytosemiotics, and so on. The obvious solution to this is built into the hierarchical nature of the phylogenetic tree. Starting at the top and working down you have domains, kingdoms, phyla, … genera, species. If there are empty shells with little research activity, one can simply go up to a higher level.

The suggested sub-fields would mirror the hierarchical structure of the phylogenetic tree at whatever level seems appropriate. Sebeok (1997) has argued for using the five kingdoms of life as proposed by Whittaker (1969). But one does not have to stick to the kingdom level or any other given level in the taxonomic hierarchy. One can go up or down in the hierarchy depending on how specific one wants to be. Thus, there is no full outline of suggested subfields because one does not have to slavishly adhere to any given level. My recommendation however, would be to start at the highest level and work down depending on the degree of detail one is willing to look at. This approach would suggest putting a lot of effort into Queiroz et al.’s (2011) second semiotic branch: “(2) cellular and molecular semiotics.”

6 Information theory, meaning, and substrate-independent minds

Claude Shannon was an early pioneer of information theory (Shannon 1948; Shannon and Weaver 1949). However, he was concerned only with the efficiency of sending signals. For biosemiotics and biology in general, sending signals efficiently is important. But of much more importance is the meaning of those signals. Signals without meaning can be sent efficiently but they play no role in biology. The only important feature of biological signals is their meaning. Information theory has almost nothing to say about meaning. Shannon was not trying to figure out if the signals had any meaning. He was just trying to figure out how to send a signal efficiently whether it was meaningful or not. Thus, a pure Shannon information theorist analyzing signals without regard to meaning, cannot be a biosemiotician.

A world from which meaning has been banished is not a biological world. The reason for this is that evolution – the process that has produced all of biology – is all about survival or persistence (Neto and Doolittle 2023). Emotions and feelings that promote survival –are felt to be meaningful. Eat. Stay healthy. Survive. Reproduce.

I have been arguing for the importance of phylogeny; the importance of evolutionary peculiarities and the contingencies of its history. Biology is a historical science. The structure and information processing of a brain depends completely on the peculiarities of selection (Figure 6). However, a significant fraction of physicists and computer scientists think they can abstract information processing out of the biological context in which it evolved. They think information processing is substrate-independent. Example:

Computation, intelligence and consciousness are patterns in the spacetime arrangement of particles that take on a life of their own, and it’s not the particles but the patterns that really matter! Matter doesn’t matter. (Tegmark 2017)

If consciousness is the way that information feels when it’s processed in certain ways, then it must be substrate-independent; it’s only the structure of the information processing that matters, not the structure of the matter doing the information processing. (Tegmark 2017)

This view is the opposite of the one I have been defending. I would argue that information processing, or the consciousness of any biological organism is phylogeny-dependent. It is survival and selection-dependent; historical-lineage dependent.

Tegmark refers to the “certain ways” in which information is being processed. I would argue that these “certain ways,” down to the tiniest detail, are and always have been historically dependent on the selection of ancestors. Brains, like all other organs, have been selected to keep the organism alive.

The meaning of all information processing – the reason it is being done – is that it has promoted the survival of the ancestors of the organism doing the processing. You can’t have a view except from a viewpoint. And all viewpoints are the result of biological evolution.

The analogous statement in computer science is: No software without hardware. When the need to survive is removed, the purpose is removed. In computerese, when the need to survive is removed something more fundamental than any hardware or software has been removed. The holisitic “utility function” has been removed. The whole reason for the existence of the computer has been removed. Missing or inappropriate utility functions cause the alignment problem graphically illustrated by Bostrom’s (2003) “Paperclip Maximizer”; a superintelligence whose only goal is to make paperclips. It fulfills its goal by turning the whole world, including all humans, into paperclips (Harari 2024: 266). The replacement of survival with some supposedly rationally chosen “utility function,” will almost inevitably lead to unanticipated consequences – the consequences that have been weeded out by 4 billion years of evolution.

Context is everything. Brains evolved to help an organism move (Roth 2013). If you remove (or abstract away) a functioning brain from its body, you are removing the brain’s reason for being. You are taking the evolutionary selection for survival of an organism and pretending that the resulting brain is background-, history- and lineage-independent. I think the result will not be a substrate-independent consciousness but an organ without a meaning – an organ reduced to a purposeless form of insanity – like taking the river away from a waterfall, or taking convection out of a thunderhead. Even for the most abstract biosemioticians, taking the meaning out of the signal is a dubious path. The assumption underlying the phylosemiotic hypothesis is that the style and goal and substrate of communication as well as the information processing, depend on the evolutionary path of the communicators. Communication is lineage dependent.

And yet many scholars share Tegmark’s (2017) opinion that: “Intelligence doesn’t require flesh, blood or carbon atoms.” Similarly, maybe communication doesn’t require flesh, blood or carbon atoms. For example, computers communicate with other computers. Doesn’t that mean communication is substrate-independent and that the phylosemiotic hypothesis is of little use, and that biosemioticians can ignore evolution? Once computer communication is no longer controlled by biologically evolved organisms, the reason for the communication will be the survival of the computer (independent of the selection by humans). Then it will be the branches of the phylogeny of the computers that bio(?)semiotics should be scaffolded upon.

The phylosemiotic hypothesis is a difficult challenge for biosemiotics. However, it seems appropriate that the boundaries of biosemiotics are as hard to define as the boundaries of signaling and communication. Using phylogeny as a scaffold would shift the focus of biosemiotics towards the fertile boundaries of the field. The alternative is to keep ignoring the wide diversity of ways in which life forms communicate with each other.

The opening up of biosemiotics could also contribute to the search for extraterrestrial intelligence (Dunér 2017), since our efforts to understand and communicate with other species on Earth (e.g., Sharma et al. 2024) are probably the best preparation for understanding and communicating with extraterrestrials (Doyle et al. 2011).

7 Summary

All biosemiosis depends on the evolution of the physiology of the organisms that produce, receive and interpret signs. An important way to understand the variety of biosemiosis is its link with phylogeny, which may be the most important tool in understanding the evolution of life on Earth. I argue that (i) the sub-fields of biosemiosis can be associated with the fundamental phylogenetic branches in the tree of life and that (ii) biosemiotics as a field, should have a hierarchy that mirrors the hierarchy of life. To paraphrase Dobzhansky (1973): Nothing in biosemiotics makes sense except in the light of evolution.