A Theory of Evolution That Looks Exactly Like Engineering

In a 2016 paper in the journal Biosemiotics, researcher Alexei A. Sharov attempts to rescue evolutionary theory from the simplistic, gene-centric model of the 20th-century Modern Synthesis. He proposes a sophisticated “informational concept of evolution” built on the idea of active, goal-directed “agents” that use “functional information” to survive and reproduce. Proponents of evolution might see this as a necessary update, providing a more robust framework for the creative power of natural processes. However, a careful analysis reveals that Sharov’s model does not explain how unguided evolution could produce complex life. Instead, it provides a detailed vocabulary for describing the operations of a pre-existing, intelligently designed system.

The Stated Goal: Fixing a Flawed Theory

Sharov rightly identifies the critical failures of standard evolutionary theory. He argues that the view of organisms as passive vessels for genes, blindly filtered by the environment, is inadequate. This old model, he notes, cannot explain the active, problem-solving behavior of living things, the complexity of heredity beyond DNA, or the origin of truly novel functions. As he puts it, this view “contradicts the very existence of humans who are active and intentional.”

To solve this, Sharov sets out to build a new framework based on the Extended Evolutionary Synthesis (EES). His paper aims to re-center evolution on the concept of “functional information,” which he defines as “a network of signs… that are used by agents to preserve and regulate their functions.” He analyzes evolution through three key processes:

1. Preservation: The active maintenance and inheritance of functional information, which includes not just copying DNA but also constructing the machinery needed to interpret it.
2. Advance: The improvement of existing functions through “selective reproduction,” where active agents make choices that influence their survival.
3. Emergence: The origin of new functions, which he argues is driven by agents reinterpreting existing information and repurposing existing structures for new, goal-directed purposes.

Sharov’s stated intent is to offer a more nuanced and powerful version of evolutionary theory, one that acknowledges the dynamic, information-rich, and semiotic (meaning-based) nature of life.

From Evolution to Engineering by a Change of Words

While the paper critiques the flaws of the Modern Synthesis, its proposed solutions do not support an unguided process. Instead, Sharov’s terminology systemically replaces the language of chance and necessity with the language of foresight and engineering.

First, his entire framework rests on “agents,” which he defines as “systems with spontaneous activity that select actions to pursue their goals.” This is a foundational principle of engineering, not unguided nature. Matter and energy do not have goals; agents do. By starting with goal-seeking agents, he assumes the very thing that an unguided process must explain. The central question—how did non-agentive matter become agentive?—is never addressed.

Second, the concept of “functional information” is a perfect description of an engineered control system. Sharov emphasizes that this information is not reducible to physics; it belongs to the realm of semiotics (meaning) and practice (use). This is precisely the distinction between the ink on a page and the message it carries, or the silicon in a chip and the software it runs. Our uniform and repeated experience shows that functional, symbolic information arises from a mind. Sharov analyzes this information in great detail but attributes it to a mindless process, contrary to all evidence from the world we actually observe.

Third, his mechanism for the “emergence” of new features is not based on random mutations but on agent-based problem-solving. He describes animals learning new behaviors—like using limbs to swim—which then “guides” the optimization of those body parts. This is top-down, goal-directed adaptation. An agent identifies a new goal (e.g., moving through water) and repurposes existing equipment to achieve it. This is analogous to an engineer finding a new application for an existing technology and then launching an R&D project to refine it. It is the opposite of a bottom-up, random-walk process.

A Blueprint for Biological Systems

Sharov’s framework is far more powerful when understood not as a theory of origins, but as a description of a designed system’s operation. The evidence he presents points directly to a pre-programmed blueprint, not a history of unguided modifications.

His model describes a world of nested agency, from molecular complexes to cells to whole organisms, all interpreting information to perform functions. This is a classic hierarchical control system, a hallmark of sophisticated engineering. His distinction between “protosemiosis” (direct command-action links in cells) and “eusemiosis” (object-oriented processing in a brain) mirrors the difference between low-level machine code and a high-level, object-oriented programming language. Both are engineered solutions for information processing at different scales of complexity.

Furthermore, his observation that the “simple” unicellular choanoflagellate Monosiga brevicollis has a genome “amazingly similar to complex multicellular organisms” is a major problem for the evolutionary narrative. Why would a “primitive” microbe evolve thousands of complex genes for cell adhesion and signaling that it does not use? From an evolutionary perspective, this is a massive waste of resources. From a design perspective, it makes perfect sense: the designer created a library of reusable code modules (genes) and deployed them as needed across different platforms (organisms). The choanoflagellate contains the foundational code library that would be more fully implemented in later, more complex designs. This is evidence of a common blueprint, not a common ancestor.

Conclusion: The Ghost in the Evolutionary Machine

In his effort to build a better theory of evolution, Sharov inadvertently reveals why the entire project is doomed. To account for the obvious purpose and complexity in biology, he is forced to import concepts like “agents,” “goals,” “interpretation,” and “meaning.” These concepts are aliens to the world of undirected physics and chemistry; they are native to the world of mind and purpose.

The paper provides no mechanism for how unguided processes could create agents that pursue goals or generate the functional information they rely on. It begins with these features already in place. By attempting to add a layer of informational sophistication to evolution, Sharov has built a Trojan horse. Inside it are all the core principles of intelligent design. Rather than showing how molecules gave rise to man, his paper provides an excellent framework for understanding how an intelligent agent engineered the systems of life.