From: jimruttshow8596
Alicia Herrero, Professor of Philosophy Emerita at Prince George’s Community College, explores the concepts of emergence and top-down causality, particularly in her new book, Context Changes Everything: How Constraints Create Coherence [02:55:00]. Her work re-examines traditional notions of causality and the relationships between parts and wholes in complex systems.
Aristotle’s Four Causes and the Focus on Efficient Cause
Historically, prior to the late 16th century, natural philosophers considered all four of Aristotle’s causes when examining nature [03:30:00]:
- Material Cause: The raw material from which something is made (e.g., clay for a pot) [03:40:00].
- Final Cause (Teleology/Purpose): The goal or purpose for which something is made (e.g., a pot for pouring water) [03:44:00].
- Formal Cause: The essence or fundamental identity that makes something what it is (e.g., what makes a pitcher a pitcher) [03:57:00].
- Efficient Cause: The actual force or energetic exchange exerted to make something (e.g., the potter’s hands shaping the clay) [04:13:00].
With the advent of modern science, there was a heavy and accidental focus on the efficient cause, leading to the discard of formal and final causes [04:40:00]. This perspective, often termed “naive Newtonianism,” suggested that if one knew the position and velocity of everything in the universe, the future and past could be predicted with total precision [06:04:00]. However, the study of complex dynamical systems reveals that formal and final causes implicitly re-enter the picture through interactions with the environment and, in Herrero’s view, through constraints [05:13:00].
”Nothing-But-ism” and the Problem of Emergence
A lingering issue from the overemphasis on efficient cause is “nothing-but-ism,” the idea that “the whole is nothing but the sum of its parts” [07:38:00]. This view often treats emergent properties as “epiphenomena” or mere “froth” without actual causal power [07:44:00].
This limited perspective struggles with phenomena like intention, which cannot be reduced to a simple chain of efficient causes (e.g., one neuron pushing another) [08:56:00]. The denial of top-down causality, where a higher-level system influences its components, leads to absurd conclusions. For instance, a “blendered” person’s chemicals won’t reassemble to go get ice cream, but a living person’s decision to get ice cream can cause their atoms and molecules to move accordingly, without violating physics [15:50:00].
Many, including some complexity scientists, are hesitant to ascribe causal powers to emergent properties, adhering to the notion of supervenience, which implies that higher-level properties depend entirely on lower-level ones without independent causal agency [17:09:00]. This resistance has led some to embrace panpsychism (the idea that consciousness is a fundamental property of matter) to explain emergence of mind [18:35:00]. However, a complexity perspective suggests that consciousness is simply another emergent level on the stack, similar to digestion, and does not require innate properties in fundamental particles [19:41:00].
Constraints as a Reconceptualization of Causality
Alicia Herrero proposes “constraint” as a more useful concept than efficient cause, especially when dealing with complex dynamical systems [10:30:00]. Constraints are broadly defined as conditions or factors that affect the behavior or possibilities within a system [29:28:00].
She categorizes constraints into:
- Context-Independent Constraints: These set the general boundaries and inhomogeneities of possibility space, taking a system far from equiprobability (randomness) [22:42:00]. Examples include gradients, polarity, charge, and fundamental principles like the Pauli Exclusion Principle or conservation laws [28:54:00].
- Context-Dependent Constraints: These take a system far from independence by linking things together [23:41:00]. Examples include:
- Catalysts [24:48:00]
- Feedback loops [24:52:00]
- Epigenetics [25:02:00]
- Temporal Constraints: The timing of actions (e.g., kicking a swing at the right moment) [25:57:00] or sequencing (e.g., steps in a process) [27:26:00].
- Spatial Constraints: The architecture or arrangement of elements (e.g., the fulcrum and length of a seesaw, the design of a traffic roundabout) [26:28:00].
- Enabling Constraints: A specific type of context-dependent constraint that, together, achieve closure, allowing a coherent whole to emerge [27:48:00].
Higher up in the organizational stack of the universe, constraints tend to define probabilities rather than black-and-white outcomes [29:43:00]. The emergence of a coherent dynamic often involves a phase transition to a continuous (analog) function, suggesting that top-down control can be analog, like a dimmer switch, allowing for timely and sensitive adjustments (e.g., homeostasis) [29:59:00]. This contrasts with purely digital, instantaneous interactions. The brain, for instance, functions with both digital (neurons firing) and analog (conceptual recognition) aspects [32:57:00].
Identity itself can be viewed as a set of interdependent constraints, defining complex, high-dimensional spaces rather than simplistic, single-dimensional categories [34:54:00].
Illustrative Examples of Emergence and Constraints
Bernard Cells
The Bernard cell experiment demonstrates emergence through constraints [50:50:00]. Heating a pan of viscous fluid uniformly from below creates a temperature gradient (a context-independent constraint). Beyond a certain threshold of instability, any minor fluctuation is amplified, leading to the self-organization of billions of molecules into rolling hexagonal convection cells [51:20:00]. The emergent cell structure then constrains the individual molecules, making them behave as if they are aware of their neighbors [51:55:00]. This process, termed “dissipative structures” by Prigogine, shows how individually constrained interactions cross a phase transition to produce a whole that then loops back to constrain its parts [53:26:00].
Autocatalytic Networks and Biological Autonomy
Moving beyond simple dissipative structures, living systems exhibit “closure of process” and “closure of constraints” [57:12:00]. In autocatalytic and hypercycles, the constraints themselves become self-perpetuating, creating the very conditions that make them possible [57:34:00]. A prime example is a cell, where internal autocatalytic reactions are responsible for building and maintaining the semi-permeable membrane that concentrates and regulates those reactions [58:31:00]. This creates a new “code” or set of rules that govern the boundary conditions, dictating what enters and exits [01:00:10].
Scaffolding, Entrenchment, and Buffers
Constraints can also manifest as:
- Scaffolds: Temporary, external structures that guide the construction of new systems [01:04:09]. They provide temporary metastable equilibrium points from which the next step can be more easily taken, much like a ratchet [01:05:01]. Scaffolds, like catalysts, lower the activation energy for something to occur [01:06:41].
- Entrenchment: A strong constraint, particularly in social systems, that resists innovation [01:05:37].
- Buffers: Mechanisms that control the relationships between the inside and the outside of a system, often related to how long an influence lasts [01:05:52].
These forms of constraint demonstrate that phenomena can have effects without necessarily being efficient causes. The fundamental laws of physics are not violated; rather, new, more complex structures with emergent properties are built upon them [01:07:49].
Top-Down Causality Without Magic
The concept of top-down causality, where the whole influences its parts, is not magical if understood through the lens of constraints rather than efficient causes [01:09:17]. For example, a culture constrains the behaviors of individuals within it by changing the likelihood of certain actions, without being a direct, efficient force [01:09:30]. Similarly, in a Bernard cell, the emergent structure influences the probabilities of how individual water molecules move [01:10:10]. This means top-down causality does not violate physical closure or conservation of energy [01:08:35].
Many-to-One Transitions and Degeneracy
The concept of “many-to-one transitions” is crucial for understanding emergence. This refers to “multiple realizability” or “degeneracy,” where many different lower-level paths can lead to the same emergent property or function [01:13:20]. Biologists are familiar with degeneracy, such as multiple amino acid sequences producing the same protein [01:13:10].
This challenges the idea of a one-to-one correlation between mental and neural events, as suggested by some interpretations of supervenience [01:12:01]. For example, if part of the brain is damaged, another part might take over its function, demonstrating that the same higher-level function can be realized by different underlying neural patterns [01:12:40]. An economy, too, can maintain its identity despite many different configurations of its components, as long as its overarching constraint structure remains within a certain range [01:13:44].
Many-to-one transitions can be seen as analogous to dimensional reduction, where many inputs concentrate into a single decision or affordance [01:14:34]. The reverse is “pluripotentiality,” where one lower level has the potential to become different functionalities, as seen in stem cells [01:15:18]. This addresses the problem of how a mental event (e.g., an intention) can exert a continuous influence over time, even though the specific physical processes underlying it may vary [01:17:20]. This dynamic perspective is crucial for understanding identity, exemplified by the Ship of Theseus paradox, where a system can retain its identity even as all its components are replaced [01:17:46].
The Future: The 4E Approach in Cognitive Science
The 4E approach to cognitive science—Embodied, Enacted, Extended, and Embedded—aligns well with Herrero’s views on constraints and causality [01:19:19]. This approach recognizes that the mind is not merely in the brain but is shaped by the body (embodied), actively shaped by behavior within a context (enacted), extends to external artifacts and tools (extended), and is deeply integrated within a specific environment (embedded) [01:20:05]. Herrero emphasizes that these coherent dynamics (e.g., a tatami mat affording sleeping in a specific culture) emerge from an accumulation and interweaving of constraints over time, which are real, objective features of reality, not just epistemological interpretations [01:21:59].