According to Ross Ashby’s cybernetic framework, the detection of any signal—including “weak signals” or outliers—is primarily an emergent artifact of the “net” (the frame of inquiry) chosen by the observer, though it is ultimately bounded by the individual’s sensory and computational capability.
Ashby’s interpretation suggests that detection is not a passive receipt of external “truth” but a relational process defined by the following factors:
1. The Observer’s “Net”: Selection as Creation
Ashby famously argued that a “system” is not a physical thing but a list of variables selected by an observer from the infinite variety present in any material object[1].
• Eddington’s Net: He illustrated this with the analogy of a scientist throwing a net into the sea; if the net has a two-inch mesh, the scientist might conclude as a “law” that all sea creatures are larger than two inches[4].
• Boundary Judgments: In this view, what is detected as a “signal” (a relevant change) versus what is dismissed as “noise” (background variety) is a direct consequence of the observer’s boundary judgments and the variables they choose to prioritize[7].
2. Detection as a Property of the “Aperture”
Whether a signal is perceived depends on the observer’s “aperture”—their capacity to record, compare, and distinguish between states[10][11].
• Variety and Discrimination: Ashby defines variety as the number of distinguishable elements[12]. If an observer lacks the power of discrimination (sensory capability) to distinguish a small deviation from the norm, that “weak signal” effectively does not exist for that system[14][15].
• Threshold Mechanisms: Detection often involves a threshold; in large nets (like the brain), activity might stay at a “basic” state unless the input exceeds a specific value, meaning a signal must be “strong” enough to surmount the observer’s internal resistance to change[16].
3. Sensory Capability as a Hard Limit
While the “frame” defines what can be seen, individual sensory capability acts as a transducer limit.
• Bremermann’s Limit: Ashby frequently cited this physical limit to show that no human or machine can process more than a finite quantity of information (1047 bits per gram per second)[19].
• Transcomputability: If the “weak signal” is buried in a combinatorial explosion of variables, it becomes transcomputable—meaning it is physically impossible for the observer to detect it given their finite resources[21].
4. The “Capricious” System and Hidden Variables
When a system appears to behave in a “capricious” or unpredictable way (showing unexplained outliers), Ashby argues this is often because the observer’s ‘net’ has left a relevant variable unobserved[2].
• Memory as a Substitute: An observer can often “detect” a signal they cannot see directly by taking the system’s past history into account, effectively using “memory” to compensate for their limited sensory access[24].
Summary: A Relational Artifact
In Ashby’s view, detection is a relational property between the observer and the thing[29]. An outlier is only an outlier relative to the “goal subset” or “focal condition” defined by the observer’s inquiry[32].
Does this perspective—that we “create” the systems we study by choosing what to ignore—change how you think about scientific objectivity?
References
[1] Ashby Mechanisms.pdf [2] Ashby’s Cybernetic Framework_ System, Parameter, and Environment.pdf [4] Ashby 1991 - Facets of Systems Science.pdf [7] Ashby - Systems and Information.pdf [10] Ashby Mechanisms.pdf [11] Ashby-Mechanisms_of_intelligence_ocr.pdf [12] Ashby 1991 - Requisite Variety and Its Implications for the Control of Complex Systems.pdf [14] the great ashby - casti.pdf [15] Transcomputability - ashby and epistemic processes.pdf [16] Ashby Mechanisms.pdf [19] Ashby-Mechanisms_of_intelligence_ocr.pdf [21] ashby ross .pdf [24] Ashby 1991 - Facets of Systems Science.pdf [29] Ashby Mechanisms.pdf [32] Ashby Mechanisms.pdf