What the science actually shows about plant signaling, learning, and suffering — and what it means for ethics
The question of plant sentience arises frequently in conversations about veganism and animal welfare — often deployed as a rhetorical counter-argument ("but what about the suffering of plants?"). But the question deserves serious scientific engagement rather than dismissal. The last two decades have produced genuinely fascinating research on plant signaling, memory-like phenomena, and stress responses. Understanding what this research actually shows — and what it doesn't — is important for intellectual honesty and for making well-grounded ethical arguments about animal welfare.
Consensus position: Plants have sophisticated biochemical signaling systems, respond to damage and stress, and exhibit phenomena that superficially resemble learning and memory. However, they lack neurons, a nervous system, a brain, or any anatomical structure capable of generating subjective experience. The probability of plant sentience — the capacity to suffer or experience — is considered vanishingly low by the mainstream scientific community in 2025. This assessment is stable and not seriously contested in peer-reviewed neuroscience or plant biology literature.
Plants release volatile organic compounds (VOCs) when damaged, including methyl jasmonate and green leaf volatiles. Neighboring plants can detect these signals and upregulate their own defenses. This is a real, well-documented phenomenon. It is a biochemical response system — analogous to the immune system, not the nervous system.
The Mimosa pudica "habituation" experiments (Gagliano et al., 2014) showed that plants dropped their leaves less in response to repeated harmless drops over time, and retained this "habituation" for days. This is real and interesting. But habituation is the simplest form of plasticity in any system and requires no subjective experience or central nervous system. No replication has confirmed this beyond basic habituation.
Pain requires: (1) nociceptors detecting tissue damage, (2) transmission via nervous system, (3) central processing generating aversive subjective experience. Plants have chemical damage-detection mechanisms but no neurons, no nervous system, and no mechanism for generating subjective experience. Calling plant stress responses "pain" is a category error that conflates mechanism with experience.
Plants do conduct electrical signals (action potentials in phloem and other tissues). These propagate from site of damage and trigger hormonal responses. This is well-documented. However, these signals are orders of magnitude slower than neural signals, operate through completely different mechanisms, and propagate through vascular tissue without generating any centralized processing. The analogy to nervous systems is superficial.
Gagliano's research on plant habituation and apparent associative learning is interesting and contested. Even Gagliano herself does not claim plants are conscious in the sense relevant to moral consideration. Her work has faced significant methodological criticism and limited replication. The mainstream plant biology community does not accept these findings as demonstrating cognition.
The "but plants feel pain too" argument is predominantly deployed not as genuine scientific inquiry but as a rhetorical device to undermine animal welfare arguments. It is worth recognizing that:
There are genuinely interesting open questions in plant biology worth taking seriously: