Pursuing Rainbows

At one point last year, I asked my students what color play dough they wanted to make.  After a few rounds of run-off voting, they decided unanimously for rainbow.  Of course.

So we made it.  Or tried at least.  We followed our usual recipe and when we got to the point where we added food coloring, we put in all the colors we had and stirred.  The result was anything but rainbow, instead yielding a muddy brown.  “It’ll turn rainbow once we cook it!”  Several minutes of applied heat did result in a change, but rather than rainbow, we now had a shiny chocolatey brown.  “It needs to cool!  Then we’ll have rainbow!”  Four hours later, the shiny, chocolatey brown play dough remained shiny and chocolatey.

All was not lost.  Color aside, this was a great batch of play dough.  It had the perfect blend of firmness and elasticity, the type that the kids can really get to work with.  And though the color was not as expected, it still served a special purpose: one child, a dark skinned African-American marveled at how similar it looked to his complexion and cherished the opportunity to play with “Michael play dough*”.

After a few weeks, the shiny, chocolatey brown play dough slowly lost its play doughness and a new batch was called for.  Again, I put the choice of color to the children.  “Rainbow!”

“We can certainly try to make rainbow again.  Should we try the same plan or a new plan?”

The agreed that a new plan was in order and set to work considering alternatives.  A few brainstorming sessions later, we narrowed down the viable options to three: paint the brown play dough; make separate batches, each a different color, and mix after cooking; and making a single batch but adding the food coloring one at a time.  Another round of voting led them to choose the separate batch method.  And we were off!

A few hours later, we had four glorious lumps of play dough: red, green, blue, and yellow.

“Did it work?” I asked them.

“Silly, Mr. Kazzy.  We haven’t mixed them yet!  You have to mix the colors.”

And we did!  And it worked!  And just like that, a very determined group of preschoolers made rainbow play dough, a feat even their teacher presumed impossible.

I have the best job ever.

———-
* If you walk the halls and classrooms of most schools, you’ll rarely see shades of brown or black used decoratively.  You also might here teachers admonish children to use “pretty colors” in their drawings and to askew these supposedly unsavory ones.  I recognize that this is often done in pursuit of a certain aesthetic, but people ought to consider the impact such decisions and such language have on children (particularly young children) of color and their self-image. Michael was so enamored with the play dough that I let him take him home with him when we were done using it. He still talks about it to this day, a full year later.

Please do be so kind as to share this post.
TwitterFacebookRedditEmailPrintFriendlyMore options

7 thoughts on “Pursuing Rainbows

    • I tend to frame my science curriculum as more about inquiry skills than content. There is some basic scientific knowledge that a 5-year-old should have, but I think it more important they learn how to ask questions, pursue answers, test hypotheses, evaluate results, make predictions, revisit predictions, incorporate new data, etc. Basically, I try to teach them the scientific method AND encourage them to understand it as not just a way of doing science (I rarely even use the word science) but of interacting with the world.

      Unfortunately, our formal science program (taught by specialist teachers) thinks otherwise. There is a lot of rote memorization and regurgitation. A lot of watching Magic Schoolbus videos. I’m considering pulling my kids from the program because I think it wastes a valuable hour of their week.

      Report

      • There is some basic scientific knowledge that a 5-year-old should have

        Is there really?

        Seriously, if you were to obliterate all the scientific knowledge of all the 5-year-olds, would that have a significant impact on what they actually end up learning by the time they are adults? And if so, is that knowledge gap likely to make for worse scientists out of those who choose to go that path?

        One approach is to teach everyone everything and then one day when they are 25 and in the second half of their PhD programs tell them it is OK to try something out. Or you can establish a habit of asking questions and answering them with experiments from the very beginning. I have to admit that I don’t have any empirical evidence as to which approach is actually better, but a good number of Nobel prize winners seem to suggest the latter even though they succeeded in systems that practice the former.

        Report

      • Well, what any student at any age “should have” is a fairly messy idea. Ultimately, it is context specific. No, I don’t know if there is some absolute set of scientific knowledge a 5-year-old should have. Though I might express some concern if they didn’t generalize a certain body of knowledge based on experience by that point. This concern would be routed in more general cognitive development. E.g., if they experienced water on a regular basis but still couldn’t tell you its basic properties, there is probably something going on there. Though that isn’t necessarily related to science.

        So, no, I don’t think there are things that are absolutely necessary. Rather, there are some things that they can more or less conceptually understand and imparting those at 4 or 5 allows other things to happen at 6 and 7 and 8, etc. There are also developmental processes that are happening at this time that make certain things more accessible and thus interesting to them. Four- and five-year-olds can begin to understand that things change over time in a way they couldn’t previously. Now, this can be explored through a variety of content areas: they can observe the cycle of a tree over the course of the year, from changing colors to shedding leaves to budding to full bloom. They can also do it by caring for a pet acquired while still a baby. They can work in a garden. They can take part in composting. It is less important that they learn the precise cycle of a tree or exactly how composting works but rather that they come to understand the way in which the world changes over time, with some of these changes being cyclical and some of these changes being linear.

        Ultimately, “what” we want children to learn is ultimately context specific. Reading isn’t all that important to the few humans still living in hunter-gatherer societies. Certainly not important enough to begin teaching it around 5 or 6. But try telling a wealthy Manhattanite that their independent school-educated 6-year-old isn’t going to get reading instruction because they don’t “need” it and see how far that gets you.

        I am a progressive educator at heart. And not progressive in the wishy-washy way that term has come to be used. I am a pretty hardcore believer in Dewey’s work on experiential education with a healthy dose of Vygotskian social-construvisim in there as well.

        Report

      • Basically, I try to teach them the scientific method AND encourage them to understand it as not just a way of doing science (I rarely even use the word science) but of interacting with the world.

        First, lots of applause for you and the kids! Second, I myself struggle to draw a line between science and engineering. I would tend to put the kids’ efforts here into the category of engineering — they’ve figured out how to make rainbow play dough, but they don’t know why the second method works where the first one didn’t. That’s okay, the methodologies are similar, at least when you’re doing engineering work at your personal leading edge. Certainly there are times when the engineers have been well out ahead of the scientists. Eg, Roman engineers knew that cement made with a certain volcanic ash would set underwater, and trial-and-error experiments taught them the proper proportions for ingredients. But it was most of two millennia before chemists could explain why the cement set and those were the right proportions.

        I’m curious about how the rainbow play dough will change over time as the colored bits get smaller and smaller. Will it eventually turn brown? Let us know how that works out, please.

        Report

      • I’ll confess to not even attempting to draw a line between science and engineering. If anything, I would have thought that engineering was a subset of science, but what do I know?

        The why question is an important one. It is one I ask of my students constantly. “Why do you think that happened?” “Why do you think that?” “Why?” I probably drive them crazy with it. But I think it an important question to ask. Even if they know the what or the how of something, asking why still matters.

        As for the play dough, constant working of the play dough saw its rainbow tie-dyed natured slowly morph into a dullish purple/gray. Not nearly as vibrant as the original brown, interestingly enough. The kids still loved it and called in rainbow until they decided that all the rainbow just went away. That was enough for them.

        Report

Comments are closed.