Permaculture Garden: The Parts Worth Borrowing

Permaculture arrives bundled: some genuinely excellent design thinking, some sound soil practice, and a layer of confident claims that have never been tested anywhere near carefully enough. I have taken the first two and left most of the third, and the parts I kept have changed how my garden works more than any individual technique.

The short answer

Take three things and you have most of the value. Place plants by how often you visit them, so herbs are at the door and the compost heap is not. Catch water where it falls, which means a butt on every downpipe. Build soil by mulching rather than digging. Then stack plants at different heights to get more from the same ground. Be sceptical of anything stated as a rule without reference to your climate and soil.

Zones: the best idea in permaculture#

If you take one thing, take this. It is not horticultural, it costs nothing, and it changes whether a garden gets used.

Place things according to how often they need you.

Zone 1, at the door

Herbs, salad, anything picked daily, and the watering can. If you walk past it every day it gets looked after. If you do not, it does not.

Zone 2, a few steps further

Main vegetable beds, soft fruit, the things you visit several times a week in season.

Zone 3, the back of the garden

Fruit trees, rhubarb, the compost heap, anything checked weekly or monthly.

Zone 4 and 5, barely touched

Rough grass, a log pile, a hedge left alone, the bit behind the shed. Wildlife lives here and you do almost nothing.

I moved my herbs from the far end of the garden to three pots by the back door and they went from occasionally used to picked most days. Nothing about the plants changed.

The corollary is just as useful: put the low attention things furthest away and stop feeling guilty about them. A rough corner that you never tend is not a failure of tidiness, it is zone five doing its job.

Water: catch it where it falls#

The second idea worth taking wholesale, and the cheapest.

A roof sheds an enormous volume of water and most gardens send all of it into a drain. A water butt on a downpipe intercepts it for the price of the butt.

Beyond that, the principle is to slow water down rather than move it away. Mulch holds it in the soil. Planting on the level rather than on a slope keeps it where the roots are. Permeable surfaces let it soak instead of running off.

Two techniques get pushed hard here and both are site specific. Swales, shallow ditches on contour, are excellent on a dry slope and a way to create a bog on flat clay. Hugelkultur, mounds built over buried wood, works in some conditions and produces a mound that dries out fast and slumps for years in others. Neither is a rule.

The EPA WaterSense guidance is the sober version of the watering side of this, and it agrees with the underlying principle: deep and infrequent, held in the soil.

Soil: mulch, do not dig#

This one I am confident about because the evidence and my own beds agree.

Digging inverts soil structure, breaks up fungal networks and brings buried weed seed to the surface where it germinates. Mulching, which means laying compost on top and letting worms take it down, preserves all of that and is markedly less work.

The practical version is a fortnight of effort a year. Clear spent crops, spread five to ten centimetres of compost on the surface, and leave it. No bare soil over winter.

Where I part company with the strongest claims is on feeding. A mature perennial system genuinely needs very little. A small garden growing hungry annuals in quantity, in my experience, still benefits from compost in volume and sometimes from a liquid feed for fruiting crops. Saying so is not a failure of principle, it is what happened.

Stacking: more from the same ground#

The idea behind a food forest, at a scale that fits a real garden.

Plants occupy different heights and root depths, so the same square metre can support several if you choose them to suit each other rather than compete. In a woodland that happens naturally. In a garden you arrange it.

Canopy

One small fruit tree on a dwarfing rootstock. Apple, plum, or a trained form against a fence if space is tight.

Shrub layer

Currants, gooseberry, or a blueberry in a pot if your soil is alkaline. Happy in the dappled shade under the tree.

Herbaceous layer

Rhubarb, perennial herbs, comfrey. Comfrey earns its place: cut it three times a year and it becomes mulch or liquid feed.

Ground cover

Strawberries, creeping thyme, wild garlic in shade. Covers soil, suppresses weeds, produces something.

Climbers

A trained climbing bean or a grape up the tree or the fence behind, using the vertical space nothing else is in.

Four layers in a few square metres is realistic and productive. The full seven layer model needs land and a decade, and presenting it as achievable in a suburban garden is where a lot of permaculture writing loses me.

What I am sceptical about, and why saying so matters#

I want to be specific rather than vaguely hedging, because this is the part most permaculture writing skips.

Companion planting is a mixture. Some of it is well supported: strong smelling plants can disrupt pest host-finding, flowers do bring in predatory insects, and mixed planting genuinely makes a crop harder for a specialist pest to locate. Some of it is folklore repeated with great confidence and no evidence, and the two arrive in the same lists looking identical.

Guild planting, the idea that particular plant combinations support each other in specific measurable ways, is mostly untested in home garden conditions. It may well be right. It is not established.

And any advice that does not mention your rainfall, soil type and climate should be treated carefully, because those three decide more than any technique. A method that transformed a garden in Oregon may do nothing in Kent and actively harm one in Arizona.

Three changes for an ordinary garden#

Nothing here requires a redesign or a philosophy.

Move the herbs to the door. Ten minutes, three pots, and it will change how much you use them.

Fit a water butt to every downpipe. Fifty dollars each, and it removes most of your summer watering problem.

Stop digging. Clear, mulch with compost, leave it. Less work and better soil, which is a rare combination.

If those three go well, add a fourth: plant one small fruit tree and stack under it. That is a permaculture garden in the only sense that matters practically.

What it costs#

Three pots of herbs at the door

$20 to $45. The highest return of anything on this page.

Water butt and diverter

$40 to $90 each. Pays for itself in one dry summer and removes a chore.

Compost for mulching

$60 to $160 a year bought in bulk, or nothing if you make your own. This is the main running cost.

One small fruit tree

$40 to $120 on a dwarfing rootstock. Ask the nursery which rootstock before you buy.

Currants and strawberries for the layers

$40 to $90, and both propagate free from cuttings and runners after that.

Comfrey, three plants

$15 to $30. Cut three times a year for mulch or liquid feed. Site it carefully, because it does not move once established.

Where this needs adjusting#

On a small or shaded plot, drop the canopy layer entirely and stack in three: shrub, herbaceous, ground cover. A fruit tree that shades the only sunny bed is a net loss.

On flat heavy clay, do not dig swales. You will make a pond. Raised beds and surface mulching are the better answer on that soil.

In a dry climate, water catching moves from useful to central, and the plant palette changes completely towards drought tolerant perennials.

And if you rent, focus on the movable parts: containers at the door, a butt if allowed, mulch on the beds. Do not plant a tree you will leave behind in a year.

Where the guidance here comes from#

Zones, stacking and water catching are standard permaculture design ideas and I have found all three genuinely useful. The no-dig soil case is well supported and matches my own beds. My reservations about companion planting and guilds are exactly that, reservations, and I would rather say so than pass on claims I cannot back. For the horticulture underneath, the RHS advice pages and Garden Organic are where I check, and Garden Organic in particular is careful about separating what is demonstrated from what is believed.

The short version#

Place things by how often you visit them. Catch water where it falls. Mulch instead of digging. Stack plants in layers to get more from the same ground. Then treat every confident rule you read as a hypothesis about someone else's soil until you have tried it in yours.

Related reading#

• Food garden ideas: stacking edibles into an ornamental garden rather than a plot.
• Mulching a garden: the no dig practice, in detail.
• Composting for beginners: where the annual mulch comes from, for nothing.
• Pollinator garden: the part of companion planting that is genuinely well supported.