The journal Nature this week debates one of the most important questions of our age: how can we feed the Earth's growing population such that no-one goes hungry and nature is left with some land and water of its own?
Being a science journal, you'll not be surprised to hear that one of the things it considers necessary in this whole arena is, er, science - in particular information about land use and the potential of land for different uses.
Jeffrey Sachs, director of the Earth Institute at Columbia University in New York and an advisor to UN Secretary-General Ban Ki-moon, and a key mover in the Millennium Development Goals initiative, argues that in fact this is the only way to make sensible decisions.
Actually, to say "he argues" is to downplay authorship of the article that bears his byline, as it's co-authored by 24 people either running or closely allied to land-use research projects such as Conservation International's Tropical Ecology Assessment and Monitoring Network.
"Making the transition to healthy, equitable and sustainable agriculture is a daunting challenge. To succeed, we will need to track and understand the diverse and changing impact of farming practices."
What they're suggesting is, in essence, a global network of research centres that can provide data that governments and farmers can use to make sensible decisions about land use.
Do this, they argue, and society will have the best opportunities to get more for less; to grow more crops, raise more cattle, whatever's needed, without spilling agriculture over into yet more millions of hectares, with all the consequences for biodiversity that would imply.
What might sound rather like a thought experiment is given tangible form by another article in Nature's set, looking at land use analysis in Brazil and the benefits it has brought.
As reporter Jeff Tollefson outlines:
"Despite rising production and persistently high commodity prices since the height of the global food crisis in 2007-08, Amazon deforestation plunged to a historic low last year, nearly 75% below its 2004 peak, and some expect more good news this year.
"This trend fuels hopes that Brazil is establishing a sustainable agricultural system that will help to feed a growing world in the decades to come."
In other words, Brazilian farmers really are learning how to get more for less.
The once-destructive soya bean industry is working - at least in some regions - alongside its arch critic Greenpeace. Using satellite images, they are working out which areas are ripe for more intensive growing, taking into account issues such as the volume of water available and what else needs it.
Cattle ranching is another hugely destructive business - not least in some of the places I visited in Acre state back in 2006, where fields ran alongside every road, and white cows stood on land that used to bear virgin forest.
With a bit of scientific input, it may be possible to double the density of cattle even in these relatively infertile fields.
Whether this sort of approach is scaleable to the entire planet, as the Sachs article argues, is something of a moot point - not least because, as he also points out, every region of the world is different - varying in temperature, rainfall, soil type, and myriad other things too - and that's without chucking in the social contexts.
There's another aspect to this that isn't tackled here: the requirements of nature itself, assuming that it is considered desirable to conserve nature as best we can.
In a nutshell, what do you do - what decisions do you make - when valuable agricultural land is also important ecologically?
More fundamentally, how do you know it is important ecologically? And given that ecology can be vital for effective farming, especially in developing countries, how do you make sure you manage the farming and the ecology together for maximal benefit to people, both through agriculture and through intact nature?
That complex kind of tapestry is being explored now in Tanzania, in the Valuing the Arc project. Here, local and Western specialists are mapping the Eastern Arc mountains for their "ecosystem value" - for what they bring through their ecological services to human residents.
It's home to three million people, and provides them with water, fuel and timber. But the three million are expanding, which potentially means more take-up of the land for farming; yet that could rebound on the services that everyone needs, farmers included.
Throughout the millennia of history, humanity has spread its footprint organically - building settlements on rivers, clearing forest for crops, moving into new areas as we expanded.
The vision presented in these papers and these projects is of a step-change in our relationship with the land - the beginning of an era where we zone our global development just as some cities zone themselves into industrial or residential areas and so on.
It should, in theory, be a more rational way of doing things. If we want to leave some reasonable portions intact for nature and ensure that nature continues providing services we need, it may be the only way of doing things.
But someone has to pay for the science that Jeffrey Sachs and others are proposing; in a recessionary world, who's that going to be?