“Paint with only water first”

by Kim Smith

Painting with the children during my artist residency at Wildflower Montessori School in Cambridge, MA, USA. Image Credit: Mary Rockett

Three years ago, I spent a year as an artist-in-residence in a Montessori classroom of three- to six-year-olds. I was really surprised the first time I walked into the space, expecting a chaotic, messy room, run amok with screaming kids — much as I remember my own preschool experience. Instead, it was a quiet and peaceful place. The children were deeply engaged in their lessons — either sitting at small tables or on rugs — and the teacher spent more time observing than speaking. Their classroom was thoughtfully ordered with shelves of simple and beautiful learning materials that I found absolutely stunning. These materials were like little minimalist sculptures — reduced forms, devoid of extraneous color or pizazz.

During my time at the preschool, I made art alongside the children and I set up environments for them to make art. The latter is actually a very challenging idea — one which I spent six years in art school trying to figure out. How do you create the limits necessary in order to yield the greatest freedom of expression? In some of my first attempts at this, I would naively bring in paints of every color of the rainbow along with all the brushes I could find. I suppose I thought that was helpful for the kids, providing numerous options for unlimited exploration. But the result was a pile of muddy paint, most of which was all over the table, floor, and their clothes; all the colors smeared into horrible brown blobs as I frantically tried to reign in the chaos I had created.

The head of school, a wise woman with few words rarely above a whisper, said to me, “Paint with only water first.”

What?

And for me, this was at the core of a key lesson about the Montessori Method: scaffold by isolating the most basic steps that comprise a more complex skill.

In art school, I had this painting teacher who would make us paint horizontal lines with sumi ink for three hours at a time. Just lines, only in black ink, over and over, until the room was covered in sheets of striped paper. What was the value of this? Paintings of black lines, while beautiful in their own right, were not the ultimate goal. We did this for both the process of creating it, and for the future work that it would yield.

I think of this exercise in a couple different ways. First, we were learning how to control a painted line, which is a difficult thing to do, requires a lot of experience, and is relevant no matter what kind of painter you are. But the other part of this, the part that reminds me of Montessori, is the act of creating a thoughtful and meditative experience that allows one to concentrate and engage completely in a single activity. I think the space of mental clarity that this activity creates and the focus that it builds is relevant, no matter what you love to do.

It’s challenging amidst the process to see the value of this kind of work — work that is not about creating the masterpiece, but rather creating the person capable of making the masterpiece.

But children don’t really think about all this; they just dive right in.

So, in the Montessori classroom, I had the children start by painting lines of water onto flat stones. After we painted with water on rocks for a week or so, the teacher suggested I give the children one color of paint. So, the following week, I gave each of them a tiny dab of white paint, one brush, and a sheet of dark blue paper.

Their paintings were beautiful. The constraints allowed for pure, sincere expression. This allowed them to do what children do best — create without inhibition. Remember, it wasn’t their first time using a paintbrush. They had pushed their limits with the water and stones, learned the delicate control of moving a brush, and then experienced the freedom of using the paint.

(both) The children’s paintings using white paint on blue paper. Image Credit: Kim Smith

Since that time, I have learned a bit more about the Montessori Method. Currently, as a Master’s student in the MIT Media Lab, I am interested in this approach once again; only this time I’m motivated by what I do not see included in the Montessori classroom: computer science. Since the Montessori Method was developed over a century ago, it does not include computer science, and traditionally there is no technology in a Montessori classroom. Now, I’m extending the Method’s original pedagogy to include a broad definition of computer science that encompasses computational thinking and concepts. I define computational thinking as the thought processes essential to breaking down problems and representing information that form the foundational ideas for how we use computers.

I was really surprised to learn that three quarters of K-12 schools don’t include computer science in their classrooms, although 9 out of 10 parents wish that they did. There are currently more than 500,000 unfilled computer science jobs in the United States, and far too few qualified applicants. But the value in learning computer science is not just in the job market. Computational thinking is a skill of problem-solving, and can be applied across other disciplines.

“In many schools today, the phrase “computer-aided instruction” means making the computer teach the child. One might say the computer is being used to program the child. In my vision, the child programs the computer and, in doing so, both acquires a sense of mastery over a piece of the most modern and powerful technology and establishes an intimate contact with some of the deepest ideas from science, from mathematics, and from the art of intellectual model building.”
(Seymour Papert, Mindstorms, p. 5)

Children today are technologically savvy. They are computer users, but they do not necessarily understand computational concepts. We have all experienced it — seeing a two-year-old at the grocery store or on the subway, pulling up pictures on a phone, or playing games…and all the while, my brilliant mother still cannot double-click a mouse.

But I am troubled when I see children, especially young kids, absorbed in a screen world that cuts them off from the multi-sensory world around them. And, there is increasing evidence that this screen-time could be doing more harm than good. In fact, the American Academy of Pediatrics recommends that we limit digital media use for children ages 2-to-5 to no more than one hour a day. An article in Psychology Today shows that excessive screen time actually damages areas of the brain that involve emotional processing, executive attention, decision making, and cognitive control.

These negative effects of too much screen time, plus the need for computer science education — which I believe can effectively be introduced at a young age — pointed me back to the connection with the Montessori Method.

Interior of Violeta Montessori School, Cambridge, MA, USA. Image Credit: Raj Bala

The Montessori classroom emphasizes a child-driven approach — one in which learners work independently and take responsibility for what and how they learn. The materials are hands-on, sequential, and align with children’s developmental stages. In fact, research by Angeline Stoll Lillard shows how Montessori students perform better on standardized tests than their traditional school peers.

With that in mind, I see connecting the computational world with non-digital learning environments as an important and challenging opportunity, and I’ve made it the basis of my current work in the Media Lab’s Social Computing research group. Together with my collaborator, Yonatan Cohen, we are designing and developing new learning materials for teaching computer science to young children, using the Montessori-inspired approach that isolates core concepts in physical, sensorial materials. The result is a series of learning materials made with natural materials, without any technology, to demonstrate such concepts as binary counting, image representation, algorithms, data structures, Boolean logic, sorting, and patterns.

Sketches and plans for developing new computer science learning materials. Image Credit: Kim Smith

Recently, I spent some time in another Montessori classroom, this time in order to observe a teacher giving a lesson about one of these new computer science materials, the Binary Towers. They are slender vertical boxes that hold wooden balls which are stacked on top of each other.

The Binary Towers material, shown here to represent the quantity of 2 as “0010” in binary. Image Credit: Kim Smith

The towers range from 8 balls-high, to 4-high, 2-high, and 1-high boxes, to represent place value for binary counting. The boxes are either filled completely to the top, or remain completely empty. When the box is full, the lid is shut and reads “1.” When it is empty, the lid remains open and it reads “0.” Children fill up the towers to discover how to count in binary, so that the lids of the towers read back the binary number for the corresponding quantity of balls.

It is easier to show than to tell, and that’s the point. The whole sensorial experience is very important — the way that the 8-balls tower is twice as tall as the 4-balls tower, which is twice the height of the 2-balls tower, and so on. The physical experience of filling the box one ball at a time is significant in that the experience of discovering binary numbers is embedded in the physical construction of these ideas.

The Binary Towers in the classroom at Violeta Montessori School in Cambridge, MA, USA. Image Credit: Kim Smith
Teacher Kari Frentzel with students at Wild Rose Montessori School in Cambridge, MA introducing a lesson about binary numbers with the Binary Towers. Image Credit: Kim Smith

Perhaps the most interesting takeaway from my observations in the classroom of the Binary Towers lesson was this: young children seem to pick up binary counting much more quickly than adults do. Certainly much faster than I did.

There are other materials in this series. The Pixel Boards consist of tile mosaics on cork boards, and enable children to explore how images are constructed through numerical codes, and how tiles represent pixels — the more pixels you have, the more detailed your image will be.

Pixel Boards in the classroom at Violeta Montessori School in Cambridge, MA, USA. Image Credit: Kim Smith
An early lesson with the Pixel Boards, in which 3–4-year-olds complete a pattern based on a matrix. Image Credit: Kim Smith

The Programming Board uses physical blocks to create simple lines of code to create drawings.

The Programming Board. Image Credit: Kim Smith

The Logic Gates demonstrate Boolean logic as mechanical operations with a marble running through it, seen here with AND gates and OR gates.

The Logic Gates, which represents Boolean logic seen here with AND and OR gates. Image Credit: Kim Smith

In her book, Montessori: The Science Behind the Genius (2005), Angeline Stoll Lillard outlines evidence that supports a link between cognition and physical movement. In younger developmental stages, that connection has been shown to improve memory, spatial representation, and verbal categorization. So when young children are engaging with materials in which they grasp, hold, and feel, they’re establishing physical understanding of these concepts.

Breaking down the basic concepts behind how computers work into physical representations exposes the black box nature of computers. Children may not entirely grasp the connection between computer science and what they’re doing, but they are forming a strong foundation that will put them ahead and prepare them for continued learning. If computer science is a new literacy, I think of these learning materials we are creating as the ABCs. Isolating these building blocks in discrete and simple physical forms is like painting on stones with water. My hope is that one day, when children are given more advanced opportunities in computer science — or when they are given all of the paint colors — they’ll create something truly magical.


Kim Smith is a Learning Innovation Fellow in the ML Learning Initiative and a research assistant in the Media Lab’s Social Computing research group. This post originally appeared on the Media Lab Website.