At the ASLA National conference 2026, I heard multiple conversations - and was pointed to Nicholas Anderson’s “Improvised Landscapes” - that were advocated for a planting design approach that, boiled down, sounds like this: plant a batch of the most aggressive/adapted plants available, let them duke it out on the site, garden by subtraction through weeding to whatever level you’re comfortable.
Okay. I see the appeal. I see situations in which this lofi approach is appropriate. But I also think, as planting designers, landscape architects, horticulturists and people who care, we can do better than this.
What Anderson describes is functionally what already happens with the majority of poorly designed naturalistic and native-oriented landscapes. Drive through any community where people are trying to do the right thing - bioswales, rain gardens, native-orienter plantings - and you’ll see: the most aggressive species survive and fill in whatever space is available. Anderson claims that this approach leads to beauty and ecological abundance. I can’t claim to have seen that as a typical outcome.
Anderson identifies two aspects that make up an “improvised” approach to planting: dynamic spatial qualities and designing subtractively through management.
Allowing for plants to sort out their positions dynamically - rather than as static groupings - is one of the great techniques that the naturalistic planting movement has normalized. In many situations, allowing for plants to position themselves dynamically makes sense. It wouldn’t necessarily be worth the effort to map out every microcondition in the topography of a dune: instead, plant some basic matrixes of plant communities and they’ll sort themselves out, moving to the areas they prefer.
A subtractive approach to design - where aesthetic and ecological goals are fulfilled primarily through weeding/removal - is also interesting. The catch is that it shifts the burden of design from something that happens primarily prior to installation into something that happens during the management phase of the design.
“Instead of a godlike, architectural approach to creating a design to be retained through constant maintenance, suppression, and intervention, this approach acknowledges that land is a time-based medium requiring continual improvisation—ideally guided by the adage, Yes, and … It’s less like building sandcastles, and more like working with the flow of a river.” Anderson writes - we shouldn’t create a design retained through maintenance, suppression and intervention - instead we should create through maintenance, suppression, and intervention.
Given that care and management are one of the most challenging phases of design to get clients to commit to paying for and successfully implement, I don’t see this working well outside of a botanic garden or the personal garden of a skilled and committed expert. That work requires a higher level of skill, knowledge, involvement and commitment than typical planting management.
The value I see in Anderson’s piece is to think clearly and explicitly about the relationship between the types of plant communities and the role of intervention expected in the landscapes we design. Anderson is focusing on highly competitive plants that quickly achieve their own equilibrium.
Levels of intervention should vary from project to project. Many landscape architects don’t have the knowledge or experience to be able to meaningfully describe or represent the levels of intervention required to bring the effects their drawings and writings promise. That needs to change. The popularity of planting design and site-scale work in the profession since 2020 seems be building respect for planting and knowledge of how to do it within the profession. But advocating for skilled management doesn’t mean abandoning design.
Anderson’s writing and the conversations I heard at the national ASLA convention felt to me like they’re carrying implicit and contradictory messages about intervention:
Intervention is embarrassing: if a landscape needs skilled care, perhaps we haven’t designed it ecologically enough. Just let the vigorous species duke it out and you’ll get something lively.
Intervention is everything: design decision-making is shifted to the care and management phase of a project. Responsibility for a poorly-performing design is shifted from the designer to maintenance.
So, which is it? I’m not opposed to improvisation - it’s inherent in any work with landscape. I am opposed to surrender masquerading as ecological design.