Microbial communities, species, clones, plasmids, transposons, integrons, and genes are evolutionary individuals tracing their evolutionary trajectories at different hierarchical levels (Baquero, 2011). Such trans-hierarchical network-like complexity simply eliminates the possibility of identification of simple causal structures, if they ever exist out of our ways of representation (Schrodinger, 1957). The hope of a simple answer to the classic Baconian question in science “What is the cause of … ” has no sense any more, and in fact the complex structure of biological processes constitute the major challenge for Biological Theory (Callebaut and Laubichler, 2007).

The challenge is not only to deal with quantitative integration of elements across these major hierarchical transitions in microbiology, or in biology at large (Maynard Smith and Szathmáry, 1997), but to eventually discover general principles of microbial life rather of just keep on descriptions (Westerhoff and Palsson, 2004). Such scaling-up process of understanding resembles the escalation from the forms of a language (as lexical or syntactic) toward its meaning (semantics) (Steels, 2004, 2010; Rosen, 2004). Interestingly, the trade-off between these hierarchical levels in a shared world has been defined as intelligibility in linguistic theory (Komarova and Niyogi, 2004). In his primary sense, the word intelligibility reflects the possibility of such a trans-hierarchical understanding. St. Thomas even derives the Latin word intelligere from intus legere, or “reading into”; even if the origin were inter legere, the term stresses the need of reading beyond the words and sentences, in another cognitive dimension, to reach the meaning. Accordingly to the Cato's classic sentence, “legere, et non intelligere, neglegere est,” that is, “as good not read, as not to understand.” In this work, we use the word “intelligibility” as the construction of meaningful (thinkable) models in response to the assimilation of knowledge (clear-and-distinct or fuzzy), and able to reflect to a certain extent the reality of complex natural systems, as those which are the objects of biological sciences. The conversion of data into knowledge constitutes a great challenge for future biological research (Brenner, 2010). In fact intelligibility is a prerequisite to developing modern biology grounded on a sound epistemology (Dougherty and Bittner, 2010). “Legere, et non intelligere, neglegere est”. Two thousand years after Cato, Albert Einstein formulated essentially the same idea: “Science without epistemology is—insofar as it is thinkable at all—primitive and muddled” (Einstein, 1949).

Metaphors are frequently used by scientists as “non-logic” epistemological aids to think about reality (De Man, 1978). Biologists have long made use of linguistic metaphors in describing and naming cellular processes, and in particular involving from DNA as language to genome as a “book of life.” The current questions are: (1) if these apparently immediate analogies might result in a deeper possibility of analysis of genetic-genomic structures using methods that have been developed in linguistic research; and (2) if such an analysis will enable to understand the general principles and processes of life, and even (not entering here again in the universal's problem) life itself as a intelligible entity.

The image of genome as a “book of life” has attracted popular imagination, but it is obvious that the knowledge of the entire genomic sequence of Haemophilus influenzae (Fleischmann et al., 1995) or Homo sapiens (McPherson et al., 2001; International Human Genome Sequencing Con-sortium, 2004) has not resulted in a much deeper understanding of the “life” clues of these organisms. This occurs not only because of our gaps in understanding the function of all genes and the complexities of regulatory and epigenetic interactions between genes and other meaningful sequences. Probably the human way of reading a text is simpler than the cell way of reading. Human language texts are read in one way only, sequentially and involving all characters. Genetic texts are “read” by cellular mechanisms in several different ways, each time using a different selection of the characters of the same text while skipping others. Indeed the “score of life” could be a better image of the genomic language. A score is a series of staves on which all the different instrumental and/or vocal parts of a musical work are written, one under the other in vertical alignment, so that the parts may be read simultaneously.

But in this report we consider there exist even bigger difficulties to predict how relevant is deciphering the language of genes, the “book of life.” One key epistemological problem is to discuss about our ability to clarify the possible relations between the structure of a possible language (genetic and genomic sequences) and the characteristics of life of particular organisms, which seem to be determined by this language. Obviously this is a problem of reductionism (Wimsatt, 1976)—might the understanding of life be reduced to the understanding of the genetic-genomic language?

4.0141

The order and qualities of the musical notes in the score, the grooves' irregularities in the gramophone record, in summary, the “language” from which music might be reproduced, is not music, but has an internal similarity with music. Much longer before the discovery of the genetic code, the Wittgenstein's theorem 4.0141 recalls the main structural feature of living organisms. The process of reading the score (genetic language), produces music (life); conversely, music can be converted, translated, by a “law of projection” into a musical score, and from this again music might be reconstructed. Without internal similarities, these transitions between series of objects “that at first sight seem to be entirely different” should be simply impossible.

Interestingly in music, as in life, the description of “what is said” beyond the individual sounds is obscure. As the classic question of Erwin Schrodinger what is Life? The question: what is Music? refuses precise answers. No propositions are transmitted by music to describe clear and distinct facts, able to be thought (logically considered) by human mind. There is, as in genetics, a certain “arithmetic order” of notes that is required to produce obscure final effects. In Leibnitz words: “exercitium arithmeticae occultum nesciendis se numere animi” (Leibnitii epistolae, collectio Kortholdi, ep. 154), that is, music is as an unconscious arithmetic's exercise in which mind do not know what is being counted.

Maybe one of the difficulties of thinking life using linguistic structures is the fluid, dynamic nature of life. Languages, music score or genetic-genomic sequences, are essentially static. A book, or a musical score, or a genome sequence can be indefinitely stored without any alteration, and even more, without producing any effect (except covering a small parcel of physical space). On the contrary, speech, music, or life, are essentially dynamic; without movement they ceases existing. The fact that linguistic structures “contain” potential dynamicity does not make them dynamic at all; indeed they are practically nothing by themselves. The key-fact is that between languages and dynamic phenomena an interpretative intermediary should be interposed. The music score gives rise to music only if interpreters are available, musicians (four in a string quartet) able to read the language and converting it into sounds. Indeed a music score has an ordered internal structure, for instance following the rules of harmony, but, at first sight, we could conclude that in the absence of correct interpretation, a music score cannot be differentiated from a random sequence of notes.

Let us now imagine an out-of-Earth scientist examining a music score. He has no idea about notes, instruments, or sounds, even about the existence of music at all. Probably he will be able to differentiate a music score from a random sequence of notes. Some notations (notes) are preferentially linked to other ones, some conserved and iterated sequences are recognizable, the role of black and white notes seems not identical, some occur more frequently than others when accompanying the name a particular instrument (unknown). The note's frequency per decimeter of score apparently depends on some mysterious words at the margin as “Andante scherzoso quasi alegretto,” that nevertheless might provide a “living equivalent.” He could conclude that the musical score has a linguistic structure, potentially leading to an unknown type of dynamic behavior. If the out-of-Earth scientist could had access to a high number of different scores, he could even trace different schools, authors, influences, even a history of this unknown language—and probably he will not be much far from reality. In summary, an analytical “science” of this language could be built, and that in the total absence of knowledge about the nature of music.

Now note that the mirror process of analysis is also possible. In that case our second out-of-Earth scientist is observing the performance of a music group playing the Schubert's Piano Trio in B flat, D. 898. Unfortunately, he does not know about the existence of music, as he is unable to hear any sound, but he is able to distinguish the keys, bows, and strings of the different instruments and he can precisely record any movement of the player's arms and fingers on these structures. A representation of these movements during time should produce something similar to the musical score of the Piano Trio. Indeed the precise record of these movements might substitute the musical score, and when applied to the instruments should reproduce the music. As in the previous case, a collection of this type of records could lead to tracing schools or authors, or a history, or even a science of this language—without knowing what music is at all.

But we can also conceive a third out-of-Earth scientist, able to hear the sounds and to correlate them with the instruments and the movements of the players. It might well happen that the scientist could perceive the separate sounds, but he is either unable to link them in his mind as significant ensembles (melodies), or the sounds are so different in his brain than in ours, that our harmony is totally useless for his sensibility, and out of any esthetical possibility. As in the previous cases, this scientist could be able to study the history of music, without understanding at all what music is.

The essential is to discuss if a particular sequence of written musical notations, or sounds, or hands and finger movements, has only the meaning of “music” when understood by a particular type of sensibility. Even more: we can replace the “out-of-Earth” scientists by musicians, which will be able to reproduce the music without knowing anything about its nature, and without experimenting any of the effects that music might cause in the appropriate sensibility. They are in a “Chinese room” situation, in which the (considered to be intelligent) intermediate within the closed room receives below the door messages in an unknown language, but accordingly to a set of rules, he is able to produce responses in the same unknown language (Searle, 1984). It is obvious that the music score, or the genome sequence, is totally unaware of its function in the process of life, and the same is true for other possible intermediaries, for instance, involved in protein translation. In the words of Sydney Brenner, “genomes do not contain in any explicit form anything at a higher level than genes” (Brenner, 1999), or, paraphrasing Leibnitz when describing monads, “genomes do not have windows.”

Natural science should be thinkable and speakable. Everything that can be thought at all can be thought clearly. Everything that can be said can be said clearly (4.116). If life is unthinkable, but if it had a reality, we should think on “pictures of life” using articulated propositions, which are “models of reality as we think it is” (4.01). “The proposition constructs a world with the help of a logical scaffolding, and therefore one can actually see in the proposition all the logical features possessed by reality if it is true” (4.023). Wittgenstein's propositions could be considered as something derived from the combinatorial and ordered nature of structures as the musical score (3.141), or, in the life context genomes, again “representing” the (suspected) reality, but unable to represent what they have in common with reality (4.12). The gap is maintained between the unthinkable but presumed reality (life) and the thinkable picture of it (proposition).

Life is as thinkable as music is thinkable. In both cases, there is what Wittgenstein calls a certain “experience of meaning.” Understanding life and music is to perceive “fine shades” of meaning. Intuitively both music and life seems to be meaningful, but in both cases they seem to be resistant to any “semantic” treatment. It has always been difficult to see how “meaning” could be fruitfully ascribed to music, as this notion is applied to language (Bar-Elli, 2006). Obviously the understanding of music has nothing to do with the ability to recreate in mind a memorized melody, or to foresee in the concert hall the next variation of a musical theme. We can say the same for life. We can predict, with a certain confidence, what will happen in the next step, based on our experiences, but that is not to understand—and even not to think about life. Experiences might provide a flavor of causality, following Hume, regularities in structured observations leads to expectation (Dougherty and Bittner, 2010), but that is not real understanding. We are just following something that we can only show, in a sense, as the conductor of the orchestra is showing with its baton, a kind of mixture of performance and unthinkable matter.

The continuous interplay between hierarchical levels is a trade-mark of life (Campbell, 1974). Some kind of unity based on reciprocal trans-hierarchical effects occurs there between what is represented and the representation itself. Indeed it is easy to imagine that the life of a particular bacterial organism is to a certain extent represented in the organization of ifs genome, as the music of a particular Mozart's string-quartet is represented in the organization of its musical score. But note that in both cases the representation is a “generational representation” for the represented entity, that is, the re-production of a particular bacterial organism or string quartet is entirely dependent from the representation, the genome, or the score, respectively. On the other hand, the score is meaningless in producing music in the absence of an instrument. In modern times (the origin of life might be another case), the genetic sequences (the representation) are only meaningful if a specific living system (what is represented) is present. The meaning of the representation should be perceived by the living system, which produces a biological scaffold (an instrument) at its turn generates its own representation, what is required for reproduction. The higher hierarchical level (the living system) has the lead in the process, as there is no representation without anything to be represented. That fits with the common wisdom in biology: the content and order of sequences in the genome corresponds to what has been selected by life itself, the complex and dynamic living interplay between the cell and its environment. For instance, speciation requires the dominance of “what it is represented,” the adapted phenotype, over a particular genome. Essentially genetic plasticity and modularity expresses such subordination. We found here an interesting trans-hierarchical cyclic correlation between representation and reproduction, a correlation that is produced blindly in both senses, as in the metaphor of the messages crossing a Chinese room (see above). Indeed we suspect that in ancient evolutionary times there was no difference between what was represented and the representation itself, the evolution of life consisting in digging a “major hierarchical transition” between both entities.

As any other type of knowledge, biology should be based on propositions. Only propositions have sense; only in the context of a proposition has a name meaning (3.3). Propositions do not have meaning, but sense, reflecting all possible situations they represent. Only the relations, the order matter, not the things, not the objects themselves. A proposition is the description of a fact (4.023), a fact that involves objects. The possibility of representation of the reality (the task of sciences) is based, accordingly to Wittgenstein, in the theorem (1.1) of the Tractatus: the world is constituted by the totality of facts, not of things (Theorem 1.1). The fact, if it is the case, exist as an elementary (atomic) fact (1.21-2), resulting from a (minimal) particular combination of objects (2.01). Objects are simple (2.02), elementary, fixed (2.026) entities. But the objects by themselves, outside facts, are only possibilities of facts, nothing in reality (2.011) except their forms, qualities to be part of facts (2.0141). These qualities determine the possibility of facts (2.012): “objects contain the possibility of all states of affairs” (2.014). The object is the fixed, the existent; the configuration is the changing, the variable (2.0271). In the elementary fact objects hang one in another, like the links of a chain (2.03), combined in a definite way (2.031). The way in which objects hang together in the fact is the structure of the fact (2.032). And finally: the form (of objects) is the possibility of the structure (2.033).

The half-seeing part might start when we could be able of developing more powerful model tools to run simultaneously all data and processes generated by these—omics, in a comprehensive and integrated way. For that a purpose we need “Big Science,” based on the convergent interactions among scientists of many disciplines, and not only from biology (Nurse, 2008). If we were able to represent the holistic result of such synthetic approach, we will be near something as a complex image of a living structure, able to be seen or to be showed. And more importantly, able to be “physiognomically” compared and related with other images obtained from other organisms. At this stage we should reach imageability. The more advanced part of the half-seeing part should be based on multi-hierarchical understanding of life, and the expected appearance of emergent qualities, particularly if communication strategies between levels are assured.

Modeling trans-hierarchical complex levels is certainly one of the biggest challenges we have (Campbell, 1974; Martínez and Moya, 2011). These models would eventually provide different levels of predictability. Indeed predictability is the best touchstone to validate the reality of complex models for complex systems (Martínez et al., 2007). Maybe even complex models might be able to predict just the next steps of biological processes, and only in close space and time compartments, just as meteorological predictions based on cumulative empirical observations. Also, and as we mentioned before (at least for non-atonal music) the experience allows to predict the evolution of a melody, at least for a few compasses. In any case, complex models will serve to continuously provide material to be tested and rejected when not mating with reality (Brenner, 2010). We should in any case be aware that to validate models by comparing them with “reality” might be a circular problem, as the “reality” could be only defined by models. That is why we should be able to “see” the reality, even in an obscure, fuzzy way. Paraphrasing Albert Einstein (Einstein, 1944; Dougherty and Bittner, 2011) the propositions, the rationality of the models (half-thinking) should be “firmly connected with sensory experiences” (half-seeing).

As Denis Noble says, life should be considered in a variety of levels; life is “a kind of music, a symphonic interplay between genes, cells, organs, body, and environment,” what can be only examined under the views of synthetic biology (Noble, 2006). Microbiologists are among the best placed scientists to mature these concepts, as they have daily experience of the complex interplay of genetic sequences and domains, operons, genes, proteins, macromolecular complexes, signaling networks, adaptive and regulatory functions, different classes of nested mobile genetic elements, clones, species, communities, integrated microbiotic ensembles, and microbial ecology at large. We have to deal with huge diversity of facts or pieces, constantly offering in a trans-hierarchical way new complex patterns to evolutionary processes (Baquero, 2004, 2009, 2011; Wagner, 2012a,b). In short, we hope that the future will allow scientists to cover the three phases of this epistemological process will be: legibility, imageability, and intelligibility of complex biological systems. Mixing half-thinking and half-seeing, the scientific method that might be we should apply to understand the complexities of microbial life.